UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Quality of life and needs for care of patients with schizophrenia
Meijer, C.J.
Publication date
2005
Link to publication
Citation for published version (APA):
Meijer, C. J. (2005). Quality of life and needs for care of patients with schizophrenia.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
Outpatientss with schizophrenia:
symptomm profiles and the relationship with
Qualityy of Life and Needs for Care
C.J.. Meijer
M.W.J.. Koeter
A.H.. Schene
S u m m a r y y A i m m
Too construct meaningful patient groups based on symptom patterns in a sample of outpa-tientss with schizophrenia and to compare these groups on service use, needs for care and qualityy of life.
Method d
1433 patients with schizophrenia from two catchment areas in The Netherlands were inclu-ded.. Two methods for the construction of patient groups were used: an empirical method 9 00 (cluster analysis) and a method based on a-priori defined criteria.
Results s
Thee a-priori method resulted in the best interpretable patient groups. Four groups were con-structed,, based on the presence of clinically relevant positive and / or negative symptoms. Differencess between the patient groups in terms of service use, needs for care and quality off life were generally in the expected direction; however they did not differentiate between alll four groups.
Conclusion n
Althoughh we could construct meaningful patient groups based on symptoms patterns in a samplee of outpatients with schizophrenia, differences between the groups were relatively smalll in terms of subjective outcome. Low overall variance in homogeneous samples can limitt the value and interpretability of empirical methods as cluster analysis.
6 . 11 I n t r o d u c t i o n
Thee psychopathological heterogeneity and definition of clinical subtypes in schizophrenic disorderss has continued to be a major issue in schizophrenia research. In recent schizophr-eniaa studies the focus has shifted from syndromes to dimensional approaches, focusing on prevailingg symptom patterns that may change over time during different phases of the illness. Variouss authors reported on empirically based classifications of patients with schizophrenia intoo psychopathological and/or neurocognitive categories (1,2,3,4,5,7,8). Results from these studiess do not permit unequivocal conclusions. Some report categorical distinctions between patientt clusters suggesting distinct symptom profiles (1). Others report patient clusters primarilyy differing in overall illness severity ranging from mild to severe overall sympto-matologyy (9). Where the above studies focus on empirically defined subgroups, others like Lindemayerr and Opler (10) suggest more theoretical, a-priori defined criteria (for example positivee or negative subtypes according to the PANSS) to define patient groups.
Thiss paper explores the existence of clinically meaningful patient groups within a represen-tativee sample of outpatients who have been in regular contact with mental health services overr the past months and who are generally considered to be in a stabilised phase of their illness.. Since the majority of these patients experience long-term cognitive and social di-sabilitiess in spite of regular anti psychotic treatment, increasing overall (social) functioning andd maximising levels of personal autonomy and quality of life continues to be an important aspectt of contemporary mental health care services (11,12,13).
Thereforee a second goal of this study is to evaluate patient groups in terms of functioning, qualityy of life and service-related variables such as subjective needs for care (on several life domains),, actual service use and the number of life domains for which care is provided. Inn general, patient groups with more severe symptomatology are expected to experience lowerr average levels of functioning (14), worse QoL (15,16,17,18) and more (unmet) needs forr care (19).
6 . 22 Method
Studyy sample and design
Inn a two-step procedure 143 patients with a diagnosis of schizophrenia (ICD F20) between 188 and 65 year of age were selected from the caseloads of specialist mental health ser-vicess in two regions in the Netherlands: an urban region (Amsterdam SouthEast) and a rurall region (West Friesland). To be eligible for this sample patients needed to have been inn contact with mental health services during the three-month period preceding the start of thee study. Amsterdam subjects also took part in the EPSILON study of schizophrenia (20). Furtherr details of the sampling procedure are described elsewhere (21).
I n s t r u m e n t s s
Symptomatology Symptomatology
Symptomatologyy was assessed by the extended 24-item version of the Brief Psychiatric
RatingRating Scale (22, 23). The BPRS is a semi structured interview that rates psychopathological
symptomss present during the previous four weeks on a Likert scale from 1 (not present) too 7 (extremely severe). The 4-Factor structure used in this study was based upon data of thee EPSILON study in five European countries, in which 404 patients with schizophrenia participatedd (24).
HealthHealth Related Quality of Life
Healthh related QoL was assessed by the Medical Outcome Study 36 Item Short Form
(SF-36)36) (25). This 36-item questionnaire comprises eight scales covering areas of physical and
mentall health and role problems resulting from health problems. The SF-36 has been ex-tensivelyy used and tested for its psychometric properties; the instrument also has satisfying psychometricc properties among patients with schizophrenia (21,26,27).
GeneralGeneral Quality of Life
Generall QoL was assessed by the Lancashire Quality of Life Profile (LQoLP), a structured interview,, measuring domain specific as well as overall life satisfaction (28). The adapted
Q2Q2 Dutch version of the LQoLP (29) comprises 25 items divided over 6 life domains.
NeedsNeeds for Care
Needss for care were assessed by the Camberwell Assessment of Need - European Version
(CAN-EU)(CAN-EU) (30,31) is a structured interview that measures needs for care on 22 life domains.
Forr each domain the patient is asked whether in the past four weeks on that particular life domainn there was either no problem, a problem for which adequate (formal or informal) caree was received (met need) or whether there was a serious problem (unmet need) on thee particular life domain.
OverallOverall Functioning
Too measure functioning we used the Global Assessment of Functioning Scale (GAF), or Axis VV of DSM IV, a measure of'psychological, social and occupational functioning' as rated by aa professional caregiver. In this study GAF was rated by a clinician or nurse who had been inn regular contact with the patient in the previous period.
Socio-demographicSocio-demographic variables and service use
Thee CSSRI-EU (Client Socio-Demographic and Service Receipt Inventory- European Version) (32)) recordssocio-demographic data, accommodation, employment, income,and all health, social,, education and criminal justice services received by a patient during the preceding months. .
Statistics s
Twoo methods, both based on symptomatology, are used to define patient groups. An e m -piricall approach (cluster analysis) and a more theoretical approach using a-priori defined categories. .
MethodMethod 1: cluster analysis
Clusterr analysis is used to classify people into discrete clusters. The key element of cluster analysiss is to obtain clusters with small within-cluster variation and large between-cluster variation,, This means that patients within a cluster are more similar than patients of dif-ferentt clusters. The following criteria are used to identify a satisfactory cluster solution: 1) thee number of clusters should be large enough to capture distinct and clinically meaningful subgroups,, 2) the amount of clusters should not be too extensive, as the number of patients inn each cluster must be large enough to make generalisations (a cluster should represent att least 10% of the patients) 3) clusters should make clinical sense in terms of clinical ex-periencee and the scientific literature.
AA categorical approach like cluster analysis offers the opportunity to go one step beyond dividingg patients into predominant symptom subgroups based on the most elevated trait dimension.. A disadvantage of cluster analysis is the fact that group allocation is based on averagee scores and therefore dependent on the amount of (within / between) group vari-ancee in the variables used in the analysis. Since our sample is relatively homogeneous with regardd to psychopathology (compared to patients in a psychotic phase of their illness), this mightt become an obstacle in finding clinically relevant patient clusters.
MethodMethod 2: a-priori categories
Usingg the positive and negative symptom scales of the BPRSE as presented by Ruggeri et al.. (24) four mutually exclusive a-priori patient groups were defined using the following procedure:: a patient is labelled P+ if s/he has at least one clinical significant symptom on thee positive scale (i.e. a BPRS item score of at least 4) and P- otherwise; a patient is label-ledd N+ if s/he has at least one clinical significant symptom on the negative scale and N-otherwise.. Patients with both P- and N- were allocated to the 'mild overall symptomatology' (MO)) subgroup. Patients with P+ and N- to the 'positive symptomatology' subgroup (PO); patientss with N+ and P- to the 'negative symptomatology subgroup' (NE) and patients with bothh P+ and N+ to a 'combined positive and negative symptomatology' subgroup (P+N).
Althoughh our criterion of at least one clinical significant symptom on a scale may seem rela-tivelyy lenient, one has to take into account that our sample comprises stabilised outpatients. Evenn though their average scale scores may be low suggesting only very modest pathology, patientss may actually suffer from one or more symptoms belonging to the s y m p t o m sub-scaless that are not reflected in their average subscale scores.
EvaluationEvaluation of the two methods and comparison of patient groups on other patient variables
Basedd on clinical relevance and interpretability a choice will be made either for (1) the empiricall or (2) the a-priori solution. Patient groups based on the preferred solution will be comparedd (using ANOVA) on functioning, (health related) quality of life, needs for care and servicee use. Post hoc tests and Bonferroni correction are performed to correct for multiple testing. .
6.33 Results
Comparisonn and description of t h e p a t i e n t clusters
AA four- as well as a five-cluster solution were explored first. The latter resulted in two very smalll clusters (n = 8 and n = 10 respectively); these small clusters collapsed into two of the otherr clusters when a 3-cluster solution was explored. Therefore we chose a final solution resultingg in three clusters with 38, 46 and 46 patients respectively.
Lookingg at overall results, table 6.1 summarises average scores on the BPRS subscales for thee three clusters. First, the clusters differ significantly on symptom type / severity. The firstt cluster shows lowest psychopathology on all subscales. Although differences between thee second and the third cluster are only very marginal in terms of average subscale sco-res,, patients in the second cluster appear to score somewhat higher on negative symptoms whereass the third cluster reports more positive symptomatology. The second and third clusterr differ (although significant) only marginally with regard to anxiety and depression.
TableTable 6.1 ThreeThree clusters based on a cluster analysis: comparison on BPRS-E subscale scores Clusterr 1 ( M O ) ' ( N = 3 8 ) ) Clusterr 2 ( N D A ) ' ( N = 4 6 ) ) C l u s t e rr 3 ( P A D ) * * ( N = 4 6 ) ) 9 4 4 MM ;sd) Att least 11 i t e m > 44 ) MM (SC Att least 11 i t e m > 44 («o) Att 'east 1 temm > MM ( s d ) Mult,, c o m p BPR5 5 S j D s c a i e s " " D . s c g a n i s a t i o n n N e g a t i v e e A^xx D e a r Positive e 11 18 ( 0 . 2 2 ) 1.155 ( 0 . 1 8 ; 11 .44 fO.421 11 16 ( 0 . 2 4 : 2 . 6 6 2 . 6 6 1 8 . 4 4 7.9 9 1.499 ( 0 . 3 6 ) 1.766 ( 0 . 5 7 ) 2 . 2 SS ( 0 . 7 3 ) 1.555 ( 0 . 6 0 ) 4 5 . 7 7 5 2 . 2 2 7 3 . 9 9 322 6 1.50 0 1.39 9 2 . 6 7 7 3 . 3 1 1 ( 0 . 4 1 ) ) ( 0 . 3 0 ) ) ( 0 . 9 2 J J (00 9 2 ) 3 0 . 4 4 2 0 . 7 7 7 6 . 1 1 9 7 . 8 8 1 0 . 6 6 6 2 5 . 5 0 0 299 77 1 2 9 . 3 3 3 0 . 0 0 0 0 . 0 0 0 00 00 0 . 0 0 0 l v s 2 . 3 3 l v s 2 . 33 2 v s 3 l v s 2 . 33 2 v s 3 l v s 2 . 33 2 v s 3
MeanMean scores for the four BPRS-E subscales ClusterCluster 1: 'MO' ('Mild Overall symptoms')'
ClusterCluster 2: 'NDA' ('Negative symptoms / Anxiety / Depression') ClusterCluster 3: 'PAD' ('Positive symptoms / Anxiety / Depression')
ANOVA'sANOVA's and multi comparison tests pertain to differences between the clusters on average subscalesubscale scores
Too characterize the patient clusters, we looked at homogeneity of the clusters, to ensure clinicall interpretability and relevance. One way to do this is to look at proportions of patients scoringg at least moderate psychopathology on the subscales and/or the individual symptoms belongingg to these subscales. We conclude that not only the clinical relevance of average scoress on the subscales between the patient clusters seems to be marginal, also the clusters aree not easily interpreted in terms of proportions of patients with moderate to severe psycho-pathologyy ( > 4). For example, although almost all patients in the third cluster report at least onee clinically relevant positive symptom, this also holds for more than 3 0 % in the second cluster.. Concluding: the three clusters are not easy interpreted in terms of (1) their clinical relevancee (differences in average subscale scores between the clusters are, although statis-ticallyy significant, only very marginal) and (2) they are not mutually exclusive (for example thee first cluster does not exclusively consist of patients with at most mild pathology, whereas theree are several patients in both clusters 2 and 3 who report only very mild pathology).
A-priorii categorisation of patients on basis of their positive and negative symptom patterns
Forr the construction of a-priori groups of patients we considered scores on positive and negativee subscales based on the BPRS-E factor analysis of Ruggeri et al (4). The construc-tionn of patient groups was based on the presence of at least one positive and / or negative symptomm reported as moderately severe to severe. This resulted in four patient groups: (1) aa group with no moderate to severe positive and negative symptoms ( n = 4 7 ) , (2) a group withh at least one serious negative symptom, but no positive symptoms (n = 20), (3) a group withh at least one serious positive symptom, but without negative symptoms (n=48) and (4) aa group reporting both serious positive and negative symptoms ( n = 1 5 ) .
Tablee 6.2 shows that the first group (indicated as 'MO' = mild overall pathology) reports lowestt average scores on most scales. The second group ('NE' = negative symptomatology) scoress highest on negative symptoms. Group 3 ('PO' = positive symptomatology) scores higherr than clusters 1 and 2 on positive symptoms and lower on negative symptoms. Finally groupp 4 ('N + P'= negative and positive symptoms) scores highest on positive symptoms andd higher than clusters 1 and 3 on negative symptoms. The NE, PO and N + P cluster do nott differ significantly on cognitive disorganisation. Levels of reported anxiety / depression aree significantly higher for the PO and (highest for) the N + P groups compared to the MO group.. Beside average subscale scores, mean item-scores as well as proportions of patients withh a score > 4 are given in table 2.
** o CLL 00 mm II zz o r HH M OO ^ -m -m o o o o o o T T rsi i - r--r r--r - H H v t t
> >
<rr r I N N> >
o o CJ J o o m m f f en n I D D ro o o o <r r o o o o o o,_, ,
ro o rsi i ro o CU U O O ^r r r-j j ro" "> >
^ ^
o o o o o o i n n U0 0 00 0 I D D ro o co o o o J-, J-, rv v rsi i o o,_, ,
ro o^ ^
T T ro o (71 1 ** i-H H^ ^
%r r r--o r--o p v v n n PM M o o o o P M M ( T l l "-H H T - l l PO O i / i i> >
LD D O O O O V V i n n o o ro o p v v v O O a i i o o PO O!/! !
> >
PM M ui i > >"-
1 1 O O o o o o ( T i i P>--( T i i O O n n m m o i i o o i / i i T T ro o o o f f" "
rv v i n n m m r-v v o o i / i i U I I o o o o V V i n n o o ro o o o o o m m Psj j O O ro o o o o o o o sr r P I I I D D rv v i n n ( T i i ( T i i O O rv v o o o o T T t t rvj j r---m r---m ( T i i a i i o o i / i i o o i n n o o n n CU U o o p v v m m m m rv v o o m m> >
Psj j > > —' ' O O o o o o ( T i i rv v cn n o o o o iP> > er. . o o -T T> >
l/l l > >'-' '
O O O O V V t t L/1 1 -VT T o o o o (N N T T rs| | t t T T> >
ro o in n > > T T PM M> >
O O O O O O PM M i - v v ÜO O 00 0 PO O 1-v v rn n P I I - H H P I I> >
P M M in n > > —' ' O O o o o o n n l O O r-. . r-. . i n n o o UÜ Ü o o -T T> >
m m ui i > > (N N^ ^
> >
O O O O o o i n n e i i rsi i vr r ro o ro o T T i-H H 'T T r i i o oM--<-" "
^ ^
r--m r--m en n !-». . o o i / i i o o LD D O O O O 00 0 o o r---U3 3 CO O r v v O O -T T O O O O V V r v v m m m m r--i n n rM M r-. . I-v v^ ^
ro o> >
ro" " o o o o o o „ „ sO O PO O r---i n n T T ro o --- H H ro o > > o o o o V V v t t en n tn tn r v v m m v f f o o vO O - H H T T> >
T T UI I > >'--u '--u O O V V O O CO O m m I D D i n n (N N ( T i i ( T l l - H H CO O
^^ ^
o o i n n to o" "
i n n i n n rN N m m L/l l - H H f f n n o o ( T i i O-i i rsi i m m m m i n n'-' '
^ ^
m m> >
o o o o o o L.r> > m m ir> > o o o o f-sl l r r rsl l - H H T T n n L/l l> >
o o o o o o n n iX) ) o o rsl l^ ^
^ H H T--rn n o o r - j j T T rsl l m m> >
a a o o o o i n n a i i o o n n ro o l-v v rsl l I I (M M T T rsi i vT T m m 10 0> >
O O o o o o rsi i no o r\i i v f f ro o L / l l 1--. . O O rM M TT T rsi i PO O i n n> >
O O CJ J o o P---i n n 0 1 1 vfr r ro o O O f f T U I N O H H N I D O O C 0 r \ j v ) - d r N r N,s r ^:Ö Ö OO O O O O O O óó o o o ó ö o PMM PO n"! PNI PM --c 0 11 rsl c o v r r-v t f "" PO T v f Psj i nn o a i PO OO PM co o oo o o ^ P-JJ PO PO lD i nn PM m a ivi---
1 1 P O O I D D rvl l^ ^
Pv v IC C rsl l co o o o 'TT 01 01 rvv PM r-. oo o o T TT T T PM *r*r o f 0 11 10 O ll PM OO -H coo m oo co PO O o o o o pvv i n oo o 13-- m co o O O O O PM M O O OO 01 rvv PM 00 0 coo 13-roo 0 r--. . P0 0 0 0 0 0 T T--O --O vT T I - v v
--PO O i D D i n n - H H m m i D D
---
1 1 O O rsl l T T^ ^
PO O O O-
1 1 r v v P v v i £ >> P O P vv CO rt rt rt rt 1X55 P O 00 - - I 0 0 PM M PM M U3 3 PM M rsl l P O O i n n^ ^
0 1 1 rsl l rtrt rt P | (S| ^ H ^ H ^ H T - H HH (N H rt O O O O O O O O O O O O O O O O O O P JJ p i n H I N >H O O O O O O O rsll ro vr PO rsi O O O O O OO O O O COO 0 1 PM 0 1 OO Tf P-v PO O O O O O ffiOifllDINNHOH ffiOifllDINNHOH O i H i D n U i d i r s I r t K ) ) O O O — l OO ^ ^ O O i DD O l i n 0 1 rsl r\l Qn POO s f Psj PM H3 0 1 ^ 11 c> —1 —1 > O S --- r-v io 0 rv a i 00 ro pv ro co a i ^ || ^-, H H 0 0 roo PM co xf i o pvv i n r v iTJ i D 00 0 0 0 0 oo m o i n i n o r o i nn H ( j i N n i P I ^ - i r M P s i ^ - ii ^ - . ^ ^ H r s i ^ ^ r s i ^ ^ - i - H O O O OO O -H O O r s J O O O O O O r s l o o rsll ^-1 r o -H PO ^ 1 rtrt P I P I H H H ^H rH 'R'R o d d d 0 0 0 0 o d d d POO CO T ^ CO * - i OO O oo o i n n 0 0 0 0_, ,
rsl l i n n P M M 0 0 s f f P M M s t t L D D 0 0 P v v 115 5 T T PM M O O LP1 1 ^1 1 T T r-v v 0 0 O O T T r-v v -f f 0 0 m m <n n PM M O O T T O O ro o i D D O O PM M PO O L0 0 T T O O ro o P v v T T O O in n a^ ^ i n n 0 0 PM M^ ^
m m i i > > 0 0 0 0 M M ro o v f f O O_, ,
PO O \D \D O O rsl l ro o ro o 0 0 0 0 Cl Cl 0 0 r-si i co o O O PM M l O O ( T i i PM M O O T T O O rsl l r-v v 0 0 m m i n n TT T 0 0 PO O ,—' '
ro o t t O O rt rt~~* ~~*
P 0 0 0 0 0 0 0 0 0 s D D r-v v p i i— —
ro o co o^ ^
p i i— —
m m ro o co o l D D O O m m^ ^
^ ^
co o— —
G 0 0 L O O O O p v v O O PM M PO O i n n—' '
0 0 PM M O O rg g i n n O O r-v v ( T l l l / l l O O „ , , PM M CO O v f f 0 0^ ^
ro o i l ) ) 0 0 PO O PM M .22 c CL a t —— c a t r » rara > - ^v c . oo 1 u T55 .^ ü ü (U U ra a 5 5 £J J> >
IT3 3 r n n a a^ ^
>--0 >--0 r r
< <
CJ J a a 0 0 (TJ J ra a>--1/1 1 0 0 X X
> >
LU U PO O CD D U U OJ J (iJ J C/l l CC = CL t O oo - 9 > y i < J ^^ L ? O ' J 15 L ^^ ^ ra TC .. O l != C <nn OJ j j .9 'Z'Z o c o ^__ 3 IP ojj (a ^ CJJ o uu . in — - . £ . £ l/l 13 100 O J1 P ^ E ^ - S ÏÏ =1-0 OO '</i ^ ro 2 cc S .2 -2 c £ o. C j ^ ^ E ^ C I ' U v ^ O C S J D - l n i i L 0 Q C Q L U Z Z ) ZZ > . L 0 < O L P i e ) r -nn ^ EE p^ s;; ^ p p n n! !
t/1 1— —
-3 -3 5 5 o. . E E VU U ss W P 0 | | £ £ oo ^ .. ^ ££ CL EE CQ SS c ca.. o EE « xx ï ^^ o Qjj IJ >> ^ (aa (a O ii aj ^^ c ^-11 ,7 ?? al u u oo ^ === & > 2 Cj (i55 ^ <->> O ui i -oo "o ww E c '-00 r-s C ï ï t L L QQ".. E 95 5 SiSi O i 3 ^ ^ oo ï . èè ^ Sü ü
r* * Ö' ' 3 3 tf) tf) 3 ; ; ' ' 5 5 3 3a a
VI VIComparingg both methods in terms of interpretability and clinical meaningfulness, we con-cludee that the a-priori approach of categorizing patients on basis of their reported positive andd negative symptoms seems to be better interpretable than the empirical approach by meanss of cluster analysis. We will describe the 4 patient groups accordingly in terms of socio-demographicc variables, functioning, quality of life and needs for care.
Comparisonn of t h e clusters on socio-demography and social integration variables
Tablee 6.3 shows that women are underrepresented in the NE group (only 5 % ) , whereas 6 0 %% of the patients in the N + P group (both positive and negative pathology) are women. Wee do not find any significant differences between the four groups with regard to socio 9 66 demographic and social integration variables and illness history.
TableTable 6.3 Four clinical patient groups: comparison on socio-demographic and social
integrationintegration variables
Agee (m, sd) Sexx (woman %)
Ethnicityy (White European
Illnesss History (m, sd) Numberr of prev. ac Sociall integration (%) Intimatee relationsh Helpfull friend Maritall status {%) Single/unmarried d Married d Divorced d Widow(er) ) Livingg situation (%) Alone e Partnerr / family Singlee parent Withh parents Withh others Protectedd living Other r
%) )
m. . P P Groupp 1 ( M O ) * * ( N = 4 7 ) ) 39.188 (9.56) 44.7 7 66.0 0 3.855 (4.64) 29.8 8 68.1 1 68.1 1 14.9 9 12.8 8 4.3 3 42.6 6 8.5 5 10.6 6 8.5 5 2.1 1 12.8 8 4.3 3 Groupp 2 ( N E ) * * (NN = 20) 43.477 (9.42) 5.0 0 65.0 0 3.000 (1,89) 10.0 0 68.4 4 80.0 0 10.0 0 10.0 0 0.0 0 63.2 2 0.0 0 5.3 3 5.3 3 5.3 3 21.1 1 0.0 0 Groupp 3 ( P O ) * * ( N = 4 8 ) ) 41.177 (10.14) 35.4 4 72.9 9 4.099 (4.33) 27.1 1 60.9 9 70.2 2 17.0 0 10.6 6 2.1 1 41.7 7 14.6 6 4.2 2 12.5 5 12.5 5 10.4 4 2.1 1 Groupp 4 (PP + N ) * ( N = 1 5 ) ) 42.077 (11.26) 60.0 0 80.0 0 3.333 (2.55) 21.4 4 71.4 4 66.7 7 20.0 0 13.3 3 0.0 0 35.7 7 14.3 3 0.0 0 7.1 1 0.0 0 21.4 4 7.1 1 p * * * 0.42 2 << 0.01 0.37 7 0.76 6 0.36 6 0.84 4 0.97 7 0.19 9PatientPatient groups: Group 1: 'MO' ('mild overall')'; Group 2: 'NE' (negative symptoms); Group 3: 'PO' (positive(positive symptoms); Group 4: P+O (positive and negative symptoms)
p-valuesp-values of Chi? tests pertaining to (%) differences between the patient groups
Comparisonn of t h e clusters on functioning and QoL
Inn table 6.4, functioning and QoL subscale scores that significantly distinguished between thee four patient groups are reported. Scores on overall functioning are generally in accor-dancee with BPRS scores; the MO group has the highest GAF score indicating best overall functioning,, whereas the N + P group has most impaired functioning. However, multiple comparisonn tests reveal that only differences between the MO group versus the three other groupss are statistically significant.
Withh regard to health related quality of life (SF-36) the same pattern emerges with most fa-vourablee QoL for the MO group; however of the eight SF-36 subscales, significant differences betweenn the groups are found only for three scales: general health, social functioning and
mentall health. Regarding social functioning, ratings for subjective (SF-36) and clinician-ratedd (GAF) social functioning are within the same direction for the four groups.
Inn terms of 'life satisfaction' as another operationalisation of QoL, patients in the MO group aree also more satisfied with their health, their leisure and social participation as well as their livee in general (LQoLP total score). Although group differences are generally small, patients withh both positive and negative symptoms are generally most dissatisfied with their life.
TableTable 6.4 Four clinical patient groups: comparison on functioning and (Health Related) QoL
Groupp 1 Group 2 Group 3 Group 4
( M O ) * * ** ( N E ) * * * ( P O ) * * * ( P + N ) * * * Mulp. (NN = 47) (N = 20) (N = 48) (N = 15) F p comp. 9 7 GAF F SF-36* * MCS S GH H SF F MH H LQoLP** * Total l Social l Health h 60.377 (11.72) 43.255 (10.29) 63.133 (21.53) 75,800 (22.63) 68.099 (16.69) 4.966 (0.61) 4.988 (0.81) 4.977 (0.76) 46.633 (11.13) 36.444 (11.80) 53.566 (20.58) 65.444 (24.42) 54.133 (17.08) 4.622 (0.64) 4.255 (1.17) 4.599 (0.76) 50.244 (11.95) 36.644 (9.65) 51.377 (20.15) 58.899 (26.47) 53.511 (17.08) 4.622 (0.72) 4.566 (1.11) 4.522 (0.79) 43,211 (11.37) 37.933 (11.40) 4 6 . 1 5 ( 2 2 . 0 0 ) ) 5 5 . 8 3 ( 3 3 . 3 6 ) ) 50.677 (20.71) 4.522 (0.68) 4.244 (0.97) 3.944 (0.97) 12.06 6 3.49 9 3.46 6 4.20 0 6.68 8 2.68 8 3.21 1 5.79 9 0.00 0 << 0.05 << 0.05 << 0.01 0.00 0 0.05 5 << Q.Ü5 << 0.01 l v s 2 , 3 , 4 4 l v s 3 3 n.s. . Ivs3 3 lvs3,4 4 n.s. . n.s. . Ivs3,4 4
Postt hoc multiple comparisons were only performed for LQoLP subscales in case overall significance
n n O O c c I-* * o o 01 1 I t t 5" " 3 3 i t t (/> >
s s
PostPost hoc multiple comparisons were only performed for SF-36 subscales in case overall significance
levelslevels were < the adjusted alpha level of 0.00625 "Ö
3" "
levelslevels were < the adjusted alpha level of 0.0083 to PatientPatient groups: Group 1: 'MO' ('mild overall')'; Group 2: 'NE' (negative symptoms); Group 3: 'PO' 5 .
(positive(positive symptoms); Group 4: P+O (positive and negative symptoms) PJ
(A A
< <
Servicee use and needs for care 3
Inn table 6.5 differences between the clusters with regard to service use and needs for care are o shown.. Differences in service contact between the four groups are non-significant. There is a 3 trendd towards a more intensive service contact, as well as more variability in contact frequency g off patients with their nurse / casemanager in the MO group and a lower contact frequency for ?j thee N + P group. On the contrary, patients in the N + P group more often visit their GP compared <*
fit fit
too the other groups, whereas GP visits are lowest for the MO group. 5 Thee MO group reports the lowest number of needs for care (5.2) whereas both the NE group
andd the N + P group report an average of 7.4 needs for care. Patients in the N+P group report » thee highest amount of unmet needs (m=4.0) compared to m = 1.7 for the MO group. Accor- 3_ dingly,, table 6.5 shows the individual life domains of the CAN for which significant differences j j . betweenn the four patient groups were found. The first column for each group ( 1 * ) gives the § totall number of perceived needs for care on that particular life domain (met needs as well
ass unmet needs for care). On most of these domains, patients in the MO group report lower proportionss of (unmet) needs for care compared to the other groups. The second column showss the prevalence of unmet needs in all groups. The NE groups reports most problems withh regard to self care, and the N + P group reports most problems (unmet needs) with regard
too physical health and safety to self (self damaging / suicidal behaviour). The majority of the Sj patientss in both the PO group and the N + P group report an unmet need on the domains psy- a etiologicall distress and treatment of psychotic symptoms. Finally, the third column for each »
fD D
groupp shows the 'relative proportion' of needs that are regarded by patients as unmet; these
g-proportionss refer to the likelihood (or 'risk') that a reported need on that particular domain g*
willl be regarded as unmet. ^
01 1
a. .
3; ; ' 'Ï Ï
" " O O o o6.44 Discussion
Thee first method of categorisation, cluster analysis, resulted in three groups. The first group wass characterised by low overall symptomatology; the second reported highest average scoress on negative symptoms, combined with more depression and anxiety; and the third clusterr reported highest positive symptoms, combined with a level of anxiety/ depression thatt was highly comparable to the second cluster. Other studies in schizophrenia using (patient)clusterr analysis based on symptomatology also tend to work with final cluster solutionss varying between three and five clusters. The instruments, the sample criteria (hospitalisedd patients versus outpatients in a stabilised phase of their illness) as well as g 33 sample sizes (varying from N=40 to N = 255) in these studies are highly variable however andd do not permit simple generalisations (see for example Dolfus et al (1), Lastra et al ( 2 ) , Vann der Does et al (3), Chouinard et al ( 9 ) , Oulis et al (33), Brazo et al (34) and Di Michele ett al (35). Most did find groups with either prevailing positive, negative or disorganised symptoms,, sometimes combined with 'mixed' groups; others however found patient clusters differingg on overall symptom level (from mild to severe overall pathology).
Althoughh cluster analysis is an adequate technique to construct subgroups, low overall variancee can limit the value and interpretability of the findings. This was likely to be the casee in this study, as we included a diagnostically homogeneous group of outpatients with schizophreniaa who (in general) were not in a psychotic phase of their illness. The clusters foundd in this study could not easily be interpreted, which in our view limits their clinical relevance.. Differences in average subscale scores between the clusters were, although sta-tisticallyy significant, marginal. As a consequence of the low average symptom scores, only modestt proportions of patients in all clusters reported moderate to severe pathology on the majorityy of the BPRS-items. Furthermore the patient clusters were not mutually exclusive inn terms of discriminating characteristics. The first cluster for example did not exclusively comprisee of patients with no or mild symptoms, whereas several patients in both clusters 2 andd 3 also reported no or mild symptoms. The above results make a clinical interpretation off the patient clusters difficult.
Thee second method used a-priori defined criteria with regard to symptom scores. In ac-cordancee with the construction of patient subtypes as suggested by Lindemayer et al (10) andd Opler et al (36) for the PANSS, we constructed groups on clinically relevant symptoms (BPRSS item scores of > 4 , indicating moderate to severe pathology) for treatment of patients inn a stabilised phase of their illness. We considered scoring > 4 on one positive or nega-tivee symptom as invalidating for the patient himself and as having implications for his/her t r e a t m e n tt in mental health care. We chose the alternative criterion of reporting at least 1 moderatee to severe positive and / or negative symptom and realise that we cannot speak off schizophrenia subtypes in this regard. With these restrictions in mind, we constructed 4 subgroups:: a group with no clinically relevant symptoms, and three groups with only nega-tivee symptoms, only positive and both positive and negative symptoms respectively.
Generalizabilityy of results found in this sample
Wee first want to place our sample in perspective to the total population of patients with schizophrenia.. One important inclusion criterion for participating in the study was that pa-tientss had been in regular (ambulatory) care in the past year; only a small minority had experiencedd psychotic relapse resulting in hospital admissions in the past year. About half of thee patients live on their own, a smaller part either lives in a sheltered living facility or lives withh their family. The generalizability of our results therefore does not comprise patients in intra-murall settings or patients who do not receive any kind of mental health care.
rtrt Ifl O Q If] oo o o o o oo o c c o z z + + 0. . «T T D.. L/1 OO 1 -—** .—* ^-v kDD r-- m n o m m HH ^ 1 (M NN L f l l D aa X 33 M-OO II Q-oo j
" "
L D D r» » i n n o o m m i n n'—' '
en n L D D m m_ _
^ ^
r---« 3 3 3 ~~ ' - re FF T; u o.. u O —|| " o o PP o r^ o o oo Ln L/I a i L/I LTII H rM t ff i n o LD o — oo -H rv n coo o L/1 Ln ^r n n^_^ ^
O JJ a i ( NN m mm n i L/ll O coo m LDD rn c c) )
,_, ,
( I l l H H c c H H, ,
* *
» »
n n^ ^
„ „ L O O L A A ( N N r \ i i T T r». . L D D L D D 0 1 1 r*> > o o L H H r \ l l m m m m r---0 r---0 r---0 r v j j m m r N N m m,-< <
L / l l CO O r v j j O O f - --!! "1 "O ii e ^ ii O 0t 11 S S ii E^ xx o !! Qj > '' Ï S <-i 11 7 j CJ 11 o & jj Q- QJ ii 0- "Gii ^
ii P L5 ii w s mm o i r NN ^ H OO O -- fM r^.. m c c_) )
,-H H 0) ) E E> >
r--* --*
* *
, ,
» »
n n ' ' „ „ rv v f ) ) o o o o - g --m --m r » i i o o i n n \ r r o o c j j n n O O L/1 1 i"-. . on n L / l l C l l O O L n n V D D O O L O O L n n O O O O || F f i ii S E ii ^EE -o ee s y ved ved 11 ? o '' S o ii !! ^ c ff Q . JU E E c c EE o 33 OJ , .. o> <Q <Q QQ I J 0ii ^ t 3 3 I jj QJ *-- CJ UII C rr o oo CM o m o SS - 5 LDD TJ- L/l O M / 11 t —— i n M i D 00 0 r\/ / Lnn r-. oo rv LOO .-< LO O " -- "D 33 * ro o PP * o-- O c -- rN m m m m:;; ;
m m o o T T ( I l l 0 0 0 U I I 'r-'r-^ H H L D D L[> > 1--^ --^
O O * * 11 f ; r-!! o. 2 ii ^ ^ < - > / . = = CC os £ Qjj ^ ^ i CLL o -f11 "Q c ^^ Q ï; p e t t o - S - ö ö T33 t r. '^^ -c 33 C J . -UU < -- Q- T3 ^^ > > > - ^ L/]] Q. Q . C L LO CLL I - I - I - O. 99 9 O O c c T3 3 01 1 5" " 3 3 V) )5 5
o o r r 01 1 3 3a a
3 3n n
n n
a a
Differencess b e t w e e n t h e clusters w i t h regard to f u n c t i o n i n g , QoL, needs for care andd service use
Becausee the a-priori patient group allocation was evaluated as allowing the most straight-forwardd clinical interpretability, we accordingly described these four patient groups in terms off overall functioning, QoL, service use and needs for care as well as socio-demographic characteristics.. Overall we expected clusters with more severe symptomatology to experience lowerr average levels of functioning, worse QoL and more (unmet) needs for care. With re-gardd to general and health related QoL our results are generally in the expected direction. Patientss from the OM group reported more favourable QoL compared to the other groups; QoLL with regard to general health, mental health and social functioning was lowest for the
1 0 00 group with both positive and negative symptoms. The combination of experiencing both
positivee and negative symptoms could refer to increased overall illness severity, resulting in moree impairment in overall functioning, deteriorated subjective health experiences as well ass decreased life satisfaction. However, the fact that the N + P group also suffers from more anxiety/depressivee symptoms may have contributed to their more negative evaluations of QoL.. We know from previous studies that anxiety/depression is an important predictor of subjectivee life satisfaction (17,18,37,38,39).
Whenn evaluating these expectations, we have to consider the possible influence of dif-ferentiall adaptational processes related to the illness (severity) which may affect patient's evaluationss on subjective criteria such as QoL; see for example Doyle et al (40) and Franz ett al (41). First, in accordance with Becker et al ( 1 4 ) , we may expect clusters with more severee symptomatology not only to experience lower overall functioning and QoL, but also too show more 'within group' variance, because these patients may be more prone to dif-ferentt adaptational processes. These processes like 'downward drift', social comparison, andd response shift could influence their evaluation of subjective functioning and QoL. For example,, patients who live in more deprived circumstances because of severe illness con-sequencess are likely to adapt their level of aspirations to their limited daily experience (the prototypee of the severely handicapped patient who has been living on a closed psychiatric wardd for years, isolated from society). Patients with higher levels of functioning, who are moree integrated in society on the contrary, may compare themselves to their'healthy' peers andd as a consequence evaluate their subjective life satisfaction as more negative.
Althoughh we did not find very convincing indications for differences in within-group vari-ancee between the clusters as a result of either differential adaptational processes or as a consequencee of particular delusional symptoms, the relatively high standard deviation in thee P+N group for the SF-36 subscales 'Social Functioning' and 'Mental Health' may reflect thee above described processes.
Differencess between the four groups with regard to the total level of subjective needs for caree were in line with scores on functioning (subjective as well as observer-rated) and QoL. Thee OM group reported the lowest overall number of needs for care (the number of life domainss on which a need for care exists. The average number of life domains on which a seriouss problem was reported as a consequence of insufficient care (the number of 'unmet needs')) was highest for the N + P group. The N + P group reported a substantial number of unmett needs on the domain 'physical health care' suggesting that their mental as well as theirr physical health may be more impaired. They may benefit from close monitoring of physicall complaints as well as better access to physical health care services. Treatment of psychoticc symptoms was regarded as insufficient by a majority of both the PO- and the N + P
group.. A point of discussion here is whether all these patients could actually benefit from a moree accurate type / dosage of AP medication or whether for patients in this cluster some residuall symptoms will pertain in spite of optimal medication prescriptions of now available AP's.. In the latter case, speaking of an 'unmeetable need' in staid o f ' u n m e t need' would actuallyy be more accurate. However, in accordance to anti-psychotic treatment, the majority off the patients in both the PO- and the N + P group report an unmet need with regard to psychologicall distress, suggesting that specific interventions focusing on feelings of anxiety andd depression could be beneficial and may contribute to improvement of their QoL. We foundd that patients in the NE-group seem to be in need of more help with self care and per-sonall hygiene compared to the other groups. In contrast to our expectations, no significant groupp differences were found for other basic need domains (such as accommodation, food etc.)) and daytime activities and social contacts. Although substantial proportions of patients reportedd a need on these domains (40 % or more of the patients in each of the four groups reportedd a need for care), the reported need frequency did not discriminate between the groupss as constructed for the purpose of this study.
Ann interesting point of discussion here is whether patient groups who report high propor-tionss of unmet needs on the same domains would also benefit from the same (changes in) servicee provision or specific interventions. The Camberwell Assessment of Need does not discriminatee in different types of required interventions; for that particular goal an instru-mentt such as the NFCAS (42) may provide valuable additive information.
Onee important issue related to the interpretation of the different constructs used here is thee extent to which they reflect a subjective patient- versus an objective observer rating. Forr example the ratings of symptom severity on the majority of the 24 BPRS items is based onn information given by the patient (for example regarding the presence of hallucinations), whereass other items are based on direct observations rated by the assessor (and therefore pertainingg to the observations made within the test situation itself). The GAF score on the otherr hand is based on an evaluation of overall functioning rated by the patient's own cli-niciann or key worker. This in contrast to for example QoL as measured by the LQoLP, which cann be seen as a completely subjective evaluation of the patient regarding his satisfaction withh his life in general, irrespective of the differential elements (circumstances, experiences, norms,, values and comparison processes) that may influence this evaluation.
Thee fact that the profile descriptions of the four groups are generally confirmed by both the subjectivee and the more objective ratings / evaluations confirm their validity. This holds true inn so far as they discriminate between the presence of clinically relevant symptoms versus noo or minor symptoms. Several subscales within the other instruments also discriminated betweenn the other groups (for example on the individual CAN domains), resulting in valuable informationn for those working in community mental health care.
References s
1.. Dollfus, S., Eventt, B., Ribeyre, J.M., Assouly-Besse, F., Sharp, C , Petit, M. (1996). I d e n t i f y i n g subtypess of schizophrenia by cluster analyses. Schizophrenia Bulletin. 22 (3): 545-55.
2.2. Lastra, I., Vazquez-Barquero, J.L., Herrera Castanedo, S., Cuesta, M.J., Vazquez-Bourgon, M.E.,
D u n n ,, G. (2000). The classification of first episode schizophrenia: a cluster-analytical ap-proach.. Acta Psychiatrica Scandinavica. 102 (1): 26-31.
3.. Does, A.J,, van der, Linszen, D.H., Dingemans, P.M., Nugter, M.A., Scholte, W.F. (1993). A di-mensionall and categorical approach to the s y m p t o m a t o l o g y of recent-onset schizophrenia.
JournalJournal of Nervous and Mental Disease. 181(12): 744-9.
4.. Gur, R.E., Mozley, P.D., Resnick, S.M., Levick, S., Erwtn, R., Saykin, A.J., Gur, R.C. (1991). Rela-1 0 22 tions among clinical scales in schizophrenia. American Journal of Psychiatry. Rela-148 (4): 472-8.
5.. Rohricht, F., Priebe, S. (2002). Mental Health Clinical Research Center on Schizophrenia. Acta
PsychiatricaPsychiatrica Scandinavica. 105 (4); 276-82.
6.. Goldstein, G. (1994). Neurobehavioral h e t e r o g e n e i t y in schizophrenia. Archives of Clinical
Neu-ropsychol.ropsychol. 9 (3): 265-76.
7.. Lysaker, P.H., Lancaster, R.S., Davis, L.W., Clements, C.A. (2003). Patterns of neurocognitive deficitss and unawareness of illness in schizophrenia. Journal of Nervous and Mental
Di-sease.sease. 191(1): 38-44.
8.. Suhr, J.A., Spitznagel, M.B. (2001). Factor versus cluster models of schizotypal t r a i t s . I I : rela-tionn to neuropsychological i m p a i r m e n t . Schizophrenia Research. 52 (3): 241-50.
9.. Chouinard, G., Albright, P.S. (1997). Economic and health state utility d e t e r m i n a t i o n s for schi-zophrenicc patients t r e a t e d with risperidone or halopendol. Journal of Clinical
Psychopharma-cology.cology. 17(4): 298-307.
10.. Lindenmayer, J,P., Kay, S.R., Opler, L. (1984). Positive and negative subtypes in acute schizo-phrenia.. Comprehensive Psychiatry: 25(4): 455-64.
1 1 .. Lehman, A.F. (1996). Evaluating o u t c o m e s of t r e a t m e n t s for persons with psychotic disorders.
JournalJournal of Clinical Psychiatry. Suppl 11: 61-7.
12.. Katschnig, H. ( 2 0 0 0 ) . Schizophrenia and quality of life. Acta Psychiatrica Scandinavica. Suppl,
407:407: 33-7.
13.. Koivumaa-Honkanen, H.T., V i i n a m a k i , H., Honkanen, R., Tanskanen, A., A n t i k a i n e n , R., Niska-nen,, L., Jaaskelainen, J., Lehtonen, J. (1996). Correlates of life satisfaction among psychia-tricc patients. Acta Psychiatrica Scandinavica. 94 (5): 372-8.
14.. Becker, T., Leese, M., K r u m m , S., Ruggeri, M., Vazquez-Barquero, J.L. and t h e EPSILON Study Group.. Needs and quality of life a m o n g patients with schizophrenia in five European cen-t r e s :: whacen-t is cen-the impaccen-t of global funccen-tioning scores7 Accepted for publication in Social
PsychiatryPsychiatry and Psychiatric Epidemiology.
15.. Kasckow, J.W., Twamley, E., Mulchahey, J.J., Carroll, B., Sabai, M., S t r a k o w s k i , S.M., Patterson, T.,, Jeste, D.V. (2001). Health-related quality of well-being in chronically hospitalized patients wit hh schizophrenia: comparison with matched outpatients. Psychiatry Research. 103 (1):
69-78. 69-78.
16.. Packer, S., Husted, J., Cohen, S., Tomlinson, G.J. (1997). Psychopathology and quality of life in schizophrenia.. Psychiatry Neuroscience. 22 (4): 231-4.
17.. Mechanic, D., McAlpine, D., Rosenfield, S., Davis, D. (1994). Effects of illness a t t r i b u t i o n and depressionn on the quality of life among persons with serious mental illness. Social Science
andand Medicine 39 (2): 155-64.
18.. Priebe, S., Kaiser, W., Huxley, P.J., Roder-Wanner, U.U., Rudolf, H. (1998). Do d i f f e r e n t subjec-tivee evaluation criteria reflect distinct constructs? Journal of Nervous and Mental Disease.
19.. Middelboe, T., Mackeprang, T., Hansson, L., Werdelin, G., Karlsson, H., Bjarnason, O., Bengts-son-Tops,, A., Dybbro, J., Nilsson, L.L., Sandlund, M., Sorgaard, K.W. (2001). The Nordic Studyy on schizophrenic patients living in the c o m m u n i t y . Subjective needs and perceived help.. European Psychiatry, 16 (4): 207-14.
20.. Becker, T., Knapp, M., Knudsen, H.C., Schene, A.H., Tansella, M., T h o r n i c r o f t , G., Vazquez-Barquero,, J.L, ( 2 0 0 0 ) . Aims, o u t c o m e measures, study sites and patient sample. EPSILON Studyy 1. European Psychiatric Services: I n p u t s Linked to Outcome Domains and Needs.
BritishBritish Journal of Psychiatry. Suppl. 39: sl-s7.
2 1 .. Meijer, C J . , Schene, A.H., Koeter, M.W. ( 2 0 0 2 ) . Quality of life in schizophrenia measured by thee MOS SF-36 and the Lancashire Quality of Life Profile: a comparison. Acta Psychiatrica
ScandmavicaScandmavica 105 (4): 293-300.
22.22. Ventura, J., Nuechterlem, K.H., Subotnik, K.L., G u t k i n d , D., Gilbert, E.A. ( 2 0 0 0 ) . S y m p t o m
dimensionss in recent-onset schizophrenia and mania : a principal c o m p o n e n t s analysis of the 2 4 - i t e mm Brief Psychiatric Rating Scale. Psychiatry Research. 97 (2-3): 129-35.
23.. Dingemans, P.M., Linszen, D.H., Lenior, M.E., S m e e t s , R.M. (1995). Component s t r u c t u r e of thee expanded Brief Psychiatric Rating Scale (BPRS-E). Psychopharmacology. 122 (3): 263-7.
24.. Ruggeri, M., Koeter, M., Schene, A., Bonetto, C., Vazquez-Barquero, J.L., Becker, T,, Knapp, M.,, Knudsen, H.C., Tansella, M., T h o r n i c r o f t , G., and the EPSILON Study Group ( 2 0 0 5 ) . Fac-torr solution of the BPRS-expanded version in schizophrenic outpatients living in five Euro-peann countries. Schizophrenia Research. 75 (1): 107-17.
25.. Ware, J.E., Sherbourne, C D . (1992). The MOS 3 6 - i t e m s h o r t - f o r m health survey (SF-36). I. Conceptuall f r a m e w o r k and item selection. Medical Care. 30 (6): 473-83.
26.. Tunis, S.L., Croghan, T.W., Heilman, D.K., Johnstone, B.M., Obenchain, R.L. (1999). Reliability, validity,, and application of the medical outcomes study 3 6 - i t e m s h o r t - f o r m health survey (SF-36)) in schizophrenic patients t r e a t e d with olanzapine versus haloperidol. Medical Care.
37(7):37(7): 678-91.
27.. Russo, J., Trujillo, C.A., Wingerson, D,, Decker, K., Ries, R., Wetzler, H., Roy-Byrne, P. (1998). Thee MOS 3 6 - i t e m Short Form Health Survey: reliability, validity, and preliminary findings in schizophrenicc outpatients. Medical Care. 36(5): 752-6.
28.. Oliver, J.P., Huxley, P.J., Priebe, S., Kaiser, W. (1997). Measuring the quality of life of severely mentallyy ill people using the Lancashire Quality of Life Profile. Social Psychiatry and
Psychi-atricatric Epidemiology. 32 (2): 76-83.
29.. Nieuwenhuizen, C. Van, Schene, A.H., Koeter, M.W., Huxley, PJ. (2001). The Lancashire Qual-ityy of Life Profile: modification and psychometric evaluation. Social Psychiatry and
Psychiat-ricric Epidemiology. 36 (1): 36-44.
30.. McCrone, P., Leese, M., T h o r n i c r o f t , G., Schene, A.H., Knudsen, H.C., Vazquez-Barquero, J.L., Lasalvia,, A., Padfield, S., White, I.R., Griffiths, G. ( 2 0 0 0 ) . Reliability of the Camberwell As-sessmentt of Need--European Version. EPSILON Study 6. European Psychiatric Services: Inputss Linked to Outcome Domains and Needs. British Journal of Psychiatry. Suppl. 39:
s34-40. s34-40.
3 1 .. McCrone, P., Leese, M., Thornicroft, G., Schene, A., Knudsen, H.C., Vazquez-Barquero, J.L., Tansella,, M., Becker, T, (2001). A comparison of needs of patients with schizophrenia in five Europeann countries: the EPSILON Study. Acta Psychiatrica Scandmavica. 103 (5): 370-9.
32.. Chisholm, D., Knapp, M.R., Knudsen, H.C., Amaddeo, F., Gaite, L., van Wijngaarden, B. (2000). Clientt Socio-Demographic and Service Receipt Inventory—European Version: development of ann instrument for international research. EPSILON Study 5. European Psychiatric Services: Inputss Linked to Outcome Domains and Needs. British Journal of Psychiatry. Suppl. 39: s28-33.
33.. Oulis, P., Lykouras, L., Tomaras, V., Stefanis, C , Christodoulou, G.N. (1999). Clinical h o m o g e n -e i t yy of DSM-IV schizophr-enic disord-ers. Psychopathology. 32 (4): 187-91.
34.. Brazo, P., Ribeyre, J.M., Petit, M., Dollfus, S. (1996). Identification by typological analysis of distinctt groups of schizophrenic patients, Applicability of a disorganized schizophrenia s u b -t y p e .. Encephale. 22(3): 165-74.
35.. Di Michele, V,, Bolino, F. (2004). The natural course of schizophrenia and psychopathological predictorss of outcome. A c o m m u n i t y - b a s e d cohort study. Psychopathology. 37(2): 98-104.
36.. Opler, L.A., Kay, S.R., Rosado, V., Lindenmayer, J.P, (1984). Positive and negative s y n d r o m e s inn chronic schizophrenic inpatients. Journal of Nervous and Mental Disease. 172 (6): 317-25.
37.. Fitzgerald, P.B., Williams, C.L., Corteling, N., Filia, S.L., Brewer, K., Adams, A., de Castella A.R.,, Rolfe, T., Davey, P., Kulkarni, J. (2001). Subject and observer-rated quality of life in schizophreniaa 2 0 0 1 . Acta Psychiatrica Scandinavica. 103: 387-392.
38.. H u p p e r t , J.D., Weiss, K.A., Lim, R., Pratt, S., S m i t h , T.E. (2001). Quality of life in schizophre-nia:: contributions of anxiety and depression. Schizophrenia Research 1: 171-180.
39.. Tollefson, G.D., Andersen, S.W., Tran, P.V. (1999). The course of depressive s y m p t o m s in
pre-1 0 44 dieting relapse in schizophrenia: a double-blind, randomized comparison of olanzapine and
risperidone.. Biological Psychiatry. 46 (3): 365-73.
4 0 .. Doyle, M., Flanagan, S., Browne, S., Clarke, M., Lydon, D., Larkin, C , O'Callaghan, E. (1999). Subjectivee and external assessments of quality of life in schizophrenia: relationship to i n -sight.. Acta Psychiatrica Scandinavica 99(6): 466-472.
4 1 .. Franz M., Meyer T., Reber T., Gallhofer B. ( 2 0 0 0 ) . The i m p o r t a n c e of social comparisons for highh levels of subjective quality of life in chronic schizophrenic patients. Quality of Life
Re-search.search. 9(5): 481-489.
42.. Brewin, C.R., Wing, J.K., Mangen, S.P., Brugha, T.S., MacCarthy, B. (1987). Principles and practicee of measuring needs in the l o n g - t e r m mentally ill: the MRC needs for care assess-m e n t .. Psychological Medicine. 17(4): 971-81.
