Psychiatry Research 209 (2013) 15–20

Contents lists available at ScienceDirect

Psychiatry Research
journal homepage: www.elsevier.com/locate/psychres

Therapeutic alliance in schizophrenia: The role of recovery

orientation, self-stigma, and insight
Sara Kvrgic a, Marialuisa Cavelti a, Eva-Marina Beck a, Nicolas Rüsch b, Roland Vauth c,n
a
Psychiatric University Clinics Basel, Basel, Switzerland
b
Department of General and Social Psychiatry, Psychiatric University Hospital Zurich, Zurich, Switzerland
c
Psychiatric University Clinics Basel, Department of Psychiatric Outpatient Treatment, Claragraben 95, CH-4005 Basel, Switzerland

a r t i c l e i n f o abstract

Article history: The present study examined variables related to the quality of the therapeutic alliance in out-patients
Received 23 September 2011 with schizophrenia. We expected recovery orientation and insight to be positively, and self-stigma to
Received in revised form be negatively associated with a good therapeutic alliance. We expected these associations to be
10 October 2012
independent from age, clinical symptoms (i.e. positive and negative symptoms, depression), and more
Accepted 22 October 2012
general aspects of relationship building like avoidant attachment style and the duration of treatment by
the current therapist. The study included 156 participants with DSM-IV diagnoses of schizophrenia or
Keywords: schizoaffective disorder in the maintenance phase of treatment. Therapeutic alliance, recovery
Therapeutic relationship orientation, self-stigma, insight, adult attachment style, and depression were assessed by self-report.
Goal orientation
Symptoms were rated by interviewers. Hierarchical multiple regressions revealed that more recovery
Service engagement
orientation, less self-stigma, and more insight independently were associated with a better quality of
the therapeutic alliance. Clinical symptoms, adult attachment style, age, and the duration of treatment
by current therapist were unrelated to the quality of the therapeutic alliance. Low recovery orientation
and increased self-stigma might undermine the therapeutic alliance in schizophrenia beyond the
detrimental effect of poor insight. Therefore in clinical settings, besides enhancing insight, recovery
orientation, and self-stigma should be addressed.
& 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Gunderson, 1990; Evans-Jones et al., 2009). For example, patients

Therapeutic alliance is defined as the affective and collaborative
bond existing between a therapist and his patient (Svensson and
Hansson, 1999). It has also been referred to as the therapeutic bond,
working alliance or helping alliance. The theoretical definitions of
the alliance have three elements in common: (1) the collaborative
nature of relationship, (2) the affective bond between patient and
therapist, and (3) the patient’s and therapist’s ability to agree on
treatment goals and tasks (Bordin, 1979). The quality of the
therapeutic alliance is a key predictor of adherence (Lecomte
et al., 2008) and was also found to be associated with higher
psychosocial functioning, reduced symptom severity and better
quality of life (Frank and Gunderson, 1990; Gehrs and Goering,
1994; Svensson and Hansson, 1999). Because of the consistent
association between therapeutic alliance and service engagement it
is important to identify variables that predict a good therapeutic
alliance (Gibbons et al., 2003). But building a strong therapeutic
alliance in schizophrenia may be a challenging endeavor due to the
nature of the clinical presentation of the illness (Frank and
n
Corresponding author. Tel.: þ41 61 699 25 25; fax: þ 41 61 699 25 35.
E-mail address: Roland.Vauth@upkbs.ch (R. Vauth).
may distrust or hold delusional beliefs about their therapist, and
therapists may find it difficult to empathize with patients’ unusual
experience (Evans-Jones et al., 2009). Given these difficulties it is
important to understand factors which improve or undermine
building therapeutic alliance.
Because agreement between therapist and patient on treatment
goals was found to be important for the development of a strong
therapeutic alliance (Martin et al., 2000; Webb et al., 2011)
variables undermining goal orientation of the patient may be
important to address. Besides impeding effects of depression
(Webb et al., 2011) and negative symptoms (Lysaker et al., 2011)
on goal orientation of the patients also self-stigma was identified to
undermine goal orientation in therapy as well (Corrigan et al.,
2009). In contrast, motivational aspects like a strong recovery
orientation were identified to facilitate goal orientation in therapy
(Waldheter et al., 2008; Corrigan et al., 2004a).
Self-stigmatizing means applying negative stereotypes of mental
illness to oneself (Corrigan and Watson, 2002) and it is followed by
feelings of shame and by coping strategies like secrecy and with-
drawal (Rüsch et al., 2006; Vauth et al., 2007). Further, self-stigma
undermines help seeking behavior (Vogel et al., 2006), adherence to
psychosocial treatment (Livingston and Boyd, 2010), more generally
social relationships (Yanos et al., 2008), and is a risk factor for

0165-1781/$ - see front matter & 2012 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.psychres.2012.10.009
16 S. Kvrgic et al. / Psychiatry Research 209 (2013) 15–20
psychiatric hospitalization (Rüsch et al., 2009). Because continuing
feelings of unworthiness and incompetency were found to be
associated with self-stigma as well as a demoralization in engage-
ment in therapy (the ‘why try’-effect; Corrigan et al., 2009), all these
consequences underline that self-stigma may undermine engage-
ment in therapy (Livingston and Boyd, 2010) and the building of a
strong therapeutic alliance in schizophrenia, respectively.
Recovery as a motivational process (for a review see Cavelti et al.,
2011) may promote engagement in therapeutic alliance as it is
supposed to facilitate the patients’ striving for the attainment of
individual life goals by successful therapy. Recovery orientation refers
to regaining a self-determined and meaningful life in spite of mental
illness. It might be achieved by finding hope that important life goals
can be attained, re-establishing a positive identity, developing mean-
ing in life, taking control of one’s life through individual responsibility,
spirituality, empowerment, and having supporting relationships (Chiu
et al., 2009).
Variables already found to be associated with quality of ther-
apeutic alliance in individuals with schizophrenia and other forms of
severe mental illness (SMI) were patient-related factors including
older age (Draine and Solomon, 1996), avoidant attachment style
(Dozier et al., 2001; Berry et al., 2008; Kvrgic et al., 2011) and more
prior service contact (Klinkenberg et al., 1998) as well as illness-
related factors like less severe symptoms (Frank and Gunderson,
1990; McCabe and Priebe, 2003; Lysaker et al., 2011) or higher
insight into illness (Johnson et al., 2008; Wittorf et al., 2009;
Barrowclough et al., 2010). Actually, insight in patients with schizo-
phrenia is the only variable which consistently was associated with
patient-rated therapeutic alliance in most studies (Dunn et al., 2006;
Wittorf et al., 2009; Barrowclough et al., 2010; Lysaker et al., 2011).
Insight is a multidimensional construct and it is defined as the
awareness of having a mental disorder, of specific symptoms, and
their attribution to the disorder, the awareness of social conse-
quences and of need for treatment (Mintz et al., 2003). Low insight
was also found to be linked to difficulties to form sustaining bonds
with others (Lysaker et al., 1998; Francis and Penn, 2001). Low levels
of insight are a risk factor for nonadherence to treatment, which is
associated with poor clinical outcome (Lincoln et al., 2007), but on
the other hand, high levels of insight have been linked to depression,
hopelessness, suicidal tendency as well as to lowered self-esteem
(Drake et al., 2004; Hasson-Ohayon et al., 2009; Restifo et al., 2009).
Self-stigma as a moderating variable can be decisive whether more
insight leads to better or worse outcome. On the other hand, self-
stigma can act as a mediator between insight and outcomes (Lysaker
et al., 2007; Staring et al., 2009, Cavelti et al., 2012). Finally, insight is
suggested to be positively associated with recovery orientation
(Mohamed et al., 2009).
Based on these studies, we expected lower self-stigma and higher
recovery orientation to uniquely contribute to the variance of better
quality of therapeutic alliance above and beyond of possible con-
founding variables such as younger age, clinical symptoms, avoidant
attachment style, and duration of treatment by the current therapist.
Second, because of the consistent findings of an association of insight
and therapeutic alliance, we hypothesize that insight contributes
additional explanatory power to the model of therapeutic alliance,
independently from recovery orientation and self-stigma.
2. Methods
2.1. Participants and procedure
The recruitment took place in Community Mental Health Centers (CHMC) in
the region of Basel, Switzerland, between February 2009 and March 2010. Patients
between 18 years and 65 years of age and diagnosed with schizophrenia or
schizoaffective disorder in the maintenance phase of their treatment (i.e. defined
as an absence of an acute psychotic episode including a first episode of
schizophrenia and no change of medication in the last 6 weeks) were asked for
study participation. Diagnoses were confirmed by the Structured Clinical Inter-
view for Diagnostic and Statistical Manual of Mental Disorders-IV Axis I Disorders
(Wittchen et al., 1997). After the procedure was fully explained, written informed
consent was obtained from all participants. Patients were informed that their
therapists were blind to their answers. Exclusion criteria were a primary diagnosis
of alcohol or substance dependency, an organic syndrome or a learning disability,
inadequate command of German to engage in therapy with a German-speaking
therapist, and/or unstable residential arrangements. The information of the
exclusion criteria were obtained from prior reports and prior interviews with
the therapists. For all interviewer-based rating scales, three research psychologists
(MA), who were blind to the results of self-ratings and the assessments of the
attending clinicians, were previously trained until a concordance of Cohen’s
kappa¼ 0.80 was achieved (Shrout and Fleiss, 1979). Participants received a
financial compensation of 40CHF (approximately 42 USD) in order to minimize
selection bias by a high refuser rate. Therapists were psychiatric trained nurses,
psychiatrists, and psychologists. To be able to suggest a more stable state of the
therapeutic relationship only patient–therapist pairs were included, which have
worked together more than 3 months or a longer time. The study was approved by
the local ethics committee.
2.2. Treatment
Treatment was not standardized but leaned on the suggestions of Dickerson
and Lehman’s (2011) supportive therapy. According to these authors, supportive
counseling in our clinical units includes providing reassurance, offering explana-
tions and clarification, and giving advices and suggestions. Treatment was done
within a multiprofessional team of in problem solving and behavioral skills
training trained nurses, social workers, psychiatrists and psychologists, depending
on changing treatment needs in the course of illness. The therapists focus on
current problems in everyday life functioning and persistent symptoms, assessed
pharmacological needs and concerns raised by having a persistent schizophrenia or
schizoaffective disorder. The mean case load of therapists was about 40 patients.
2.3. Measurements
All measures employed have shown to be valid and reliable in samples of
patients with schizophrenia or other severe mental illnesses in prior studies (Kay
et al., 1987; Birchwood et al., 1994; Hall, 1995; Beck et al., 1996; Corrigan et al.,
1999; Corrigan et al., 2006; McGuire-Snieckus et al., 2007; Kvrgic et al., 2011).
Measures were applied once during the ongoing therapy.
Therapeutic alliance was measured using the German version of the Scale to
Assess the Therapeutic Relationship–Patients Version (STAR-P; McGuire-Snieckus
et al., 2007). The STAR is based on the pantheoretical model of therapeutic alliance
(Catty et al., 2007) and it is a self-rating instrument with 12 items comprising
three subscales: Positive Collaboration, Positive Clinician Input, and Non-
Supportive Clinician Input. Items were rated on a 5-point Likert scale, with
0¼‘never’ to 4¼ ‘always’. Before scoring, scores for the Non-Supportive Clinician
Input subscale were reversed. A total score can be obtained by summing up the
relevant subscale scores. Higher scores denote a better alliance. In the current
study Cronbach’s alpha for the total score was 0.71. We only applied the patients’
version of STAR, as a higher predictive impact on therapy outcome was demon-
strated for patient rated alliance than it was shown for therapist rated alliance in
people with schizophrenia (Horvath and Symonds, 1991; Bentall et al., 2002).
Recovery orientation was assessed with the Recovery Assessment Scale (RAS;
Corrigan et al., 1999) which is a self-rating 5-point Likert scale with response
categories from 1¼‘strongly disagree’ to 5¼ ‘strongly agree’. A factor analysis
resulted in five factors, namely Personal Confidence, Willingness to Ask for Help,
Goal and Success Orientation, Reliance on Others, and Not Dominated by Symptoms
totaling 24 items (Corrigan et al., 2004b). A total score can be calculated by summing
up all items. In the present study, Cronbach’s alpha was¼ 0.78.
We measured self-stigma using the 10-item Self-Esteem Decrement Due to
Self-Stigma subscale of Corrigan’s Self-stigma in Mental Illness Scale (Corrigan
et al., 2006; Rüsch et al., 2006). The measure included statements such as
‘I currently respect myself less because I cannot be trusted’ and ‘I currently
respect myself less because I am unpredictable’. Research participants were asked
to respond to each item using a 9-point agreement scale (9¼ ‘strongly agree’). In
the current study, Cronbach’s alpha for the Self-Esteem Decrement Due to Self-
Stigma subscale was 0.84.
Insight was measured with the 8-item Birchwood Insight Scale (BIS;
Birchwood et al., 1994), including the subscales Perceived Need for Treatment,
Awareness of Illness, and Relabeling of Symptoms as Pathological. Items are rated
from 0¼ ‘not right’ to 2 ¼‘right’, higher sum scores indicating more insight.
Cronbach’s alpha for the total score was 0.60.
Positive and negative symptoms were assessed by the Positive and Negative
Syndrome Scale (PANSS; Kay et al., 1987), a semi-structured interview composed
by 30 items, which assesses positive symptoms, negative symptoms, and general
psychopathology. High scores indicate high levels of symptoms. In the present
 S. Kvrgic et al. / Psychiatry Research 209 (2013) 15–20
study, Cronbach’s alpha was 0.67 for the Positive- and 0.74 for the Negative-
Symptoms subscale.
Adult attachment style was measured by the Psychosis Attachment Measure
(PAM; Berry et al., 2006), a self-rating scale that measures attachment avoidance
and attachment anxiety specifically in patients suffering with psychotic experi-
ences. Items are rated on a 4-point Likert scale with response categories from
0¼ ‘strongly disagree’ to 4¼‘strongly agree’. Total scores were calculated for each
dimension by averaging individual item scores, with higher scores reflecting higher
levels of anxiety and avoidance. A high overall total score reflected a general
insecure attachment style. The measure showed good psychometric characteristics
(Kvrgic et al., 2011, Berry et al., 2006). Cronbach’s alpha reached 0.71 for the
Attachment Avoidance subscale and 0.73 for the Attachment Anxiety subscale.
Depression was assessed with the Beck Depression Inventory II (BDI-II; Beck
et al., 1996). It consists of 21 items, each with four statements indicating
increasing severity (4-point Likert scale from 0 to 3). By summing up single items,
a total score is achieved ranging from 0 to 63; a high total score indicates high
levels of depressive feelings. Cronbach’s alpha reached 0.90.
To evaluate psychosocial functioning, we applied the Modified Global Assess-
ment of Functioning (Hall, 1995). This measure assesses the individual’s overall
functioning evaluated by the therapist on a rating scale ranging from 0 to 100.
A score of 100 on the M-GAF means superior functioning whereas a score of 40 or
below means severe impairments in several areas, such as work or school, family
relations, judgment, thinking, or mood.
Some measurements were translated by the authors (STAR-P, RAS, Self-Esteem
Decrement Due to Self-Stigma subscale, BIS, PAM). The adaptation of the English
version into German was carried out according to the International Test Commis-
sion Guidelines for Translating and Adapting Tests. So our adaption process took
full account of linguistic and cultural differences among the populations for whom
the adapted version of the instrument is intended (International Test Commission
2010). The translation of the English versions of the instruments into German was
carried out according to the forward–backward procedure (Stieglitz, et al., 1998).
After a member of the research team translated the original version into German,
a bilingual PhD student translated the German version back into English without
referring to the original English instrument. The differences between the back-
translated and the original English version were minimal and the final version was
developed by consensus.
In order to avoid multicollinearity, only total scores (STAR-P, RAS, BIS, BDI-II)
or single subscales (Self-Stigma Due to Self-Esteem Decrement, Avoidant Attach-
ment Style, PANSS positive and negative symptoms) were included in the
statistical analyses.
2.4. Statistical analyses
In order to control for diagnostic category before lumping the data together, a
t-test for independent samples was performed to examine if individuals with
schizophrenia disorders differed from individuals with schizoaffective disorders
with regard to the therapeutic alliance ratings.
Before running regression analyses, prerequisites for this method were
analyzed. Subsequently, a multiple regression analysis was conducted. STAR-P
total score was entered as depended variable.
All possible confounding variables found to be associated with therapeutic
alliance were entered as a first block in hierarchical regression analyses: age,
duration of treatment by the current therapist, depression, positive and negative
symptoms, and avoidant attachment style. In a second step, recovery orientation
and self-stigma were entered together in the regression analysis. Both variables
were found to influence goal orientation of patients with schizophrenia (Corrigan
et al., 2004a, 2009) and so may have an impact on therapeutic alliance, too.
Additionally, insight was entered separately in the third step in the regression
model because it is the only variable in previous studies, which was found to be
consistently correlated with patient-rated therapeutic alliance (Wittorf et al.,
2009; Barrowclough et al., 2010; Lysaker et al., 2011). Because insight was also
found to be correlated with self-stigma (Lysaker et al., 2007; Staring et al., 2009;
Cavelti et al., 2012) and associated with recovery orientation (Mohamed et al.,
(2009)) we wanted to prove whether insight would have an additional effect on
therapeutic alliance over and above to recovery orientation and selfstigma.
All data analyses were performed using SPSS for Windows, version 19.0 (SPSS
Inc., Chicago, IL, USA). All statistical tests were two-tailed and significance levels
were set at p o0.05.
3. Results
3.1. Study sample and measures
One hundred and two participants (65.4%) were diagnosed
with schizophrenia and 54 (34.6%) with schizoaffective disorder.
There was no significant difference between the diagnosis-groups
according to the STAR-P scores (t ¼0.74, d.f.¼151, p ¼0.46).
17
Therefore, and because we did not focus on an acute phase
patient sample presuming confounding effects of strong affective
symptoms, both patient samples were taken together for the
subsequent analyses. The mean age of the patients was 44.5 years
(S.D.¼ 11.67) and 102 (65.4%) were male. The interval between
the appointments was for the majority of patients 2–3 weeks
(n ¼66; 42.3%), for a smaller proportion 1 week (n ¼27; 17.3%) or
4 weeks (n ¼27; 23.7%); and only a minority of patients meet the
therapists more frequent than once/week (n ¼9, 5.8%) or had
intersession interval longer than 4 weeks (n ¼17; 10.9%). The
majority of participants lived alone (n ¼88; 56.41%) were not in a
stable partnership—defined as lasting 3 months or longer
(n ¼119; 76.3%)—and had no children (n ¼121; 77.6%). The mean
years of education was 12.31 (S.D. ¼2.93). The majority of
participants were either unemployed (n ¼90; 57.7%) or were in
sheltered employment (n ¼48; 30.8%) and received a governmen-
tal disability annuity (n ¼127; 81.4%). The mean duration of
illness was 18.63 years (S.D. ¼11.82), mean age of illness onset
was 27.79 (S.D. ¼13.53). The average duration of treatment by the
current therapist was 4.29 (S.D. ¼4.50) years. On average, parti-
cipants had been hospitalized 8.21 times (S.D. ¼6.45) and had
received 7.12 years (S.D. ¼6.56) of outpatient treatment in our
CMHCs. The majority (n¼154; 98.7%) were treated with anti-
psychotic medication, most of them with an atypical antipsycho-
tic (n ¼93; 59.6%). The majority of the therapists were nurses
(n ¼9; 42.9%) or psychiatrists (n¼9; 42.9%) and a smaller sub-
group psychologists (n ¼3; 14.3%). Mean and standard deviation
of all measures are summarized in Table 1.
3.2. Regression analyses
In a first step, prerequisites for the regression analysis were
examined: Linearity (measured with a scatterplot) between pre-
dictors and dependent variable was given, and are available by
the first author by request. The errors for different response
variables did not show different variances (i.e. no heteroskedas-
ticity, checked with a scatterplot). We further checked on this
issue by testing normal distribution of the residuals with
Kolmogorov–Smirnoff-Test (K–S-Z¼ 1.011, p¼0.259), indicating
normal distribution of the residuals. We assessed multicollinear-
ity by examining tolerance and the Variance Inflation Factor (VIF).
Multicollinearity exists when tolerance is below 0.1 and VIF is
greater than 10 or an average much greater than 1. In our case,
there was no multicollinearity.
Table 1
Range of scores, means and standard deviation of applied measures.
Range of M (S.D.)
scores
Therapeutic alliance (STAR-P) 0–48 37.26 (7.71)
Recovery (RAS) 24–120 90.00 (14.48)
Self-esteem decrement due self-stigma 10–90 25.73 (13.23)
(SSMIS)
Insight (BIS) 0–12 9.54 (2.36)
Symptom severity (PANSS)
Positive Symptoms 7–49 13.87 (4.88)
Negative Symptoms 7–49 13.91 (4.97)
Depression (BDI-II) 0–63 11.07 (9.86)
Avoidant attachment style (PAM) 0–3 1.29 (0.57)
Social functioning (M-GAF) 0–100 49.82 (10.25)
Note: M ¼mean; S.D. ¼ standard deviation. STAR-P ¼Scale to Assess Therapeutic
Relationship-Patient Version; RAS ¼Recovery Assessment Scale; SSMIS ¼Self-
esteem decrement scale of the SSMIS; BIS¼ Birchwood Insight Scale; PANSS ¼Posi-
tive and Negative Syndrome Scale; BDI-II ¼ Beck Depression Inventory; PAM ¼
Psychosis Attachment Measure; M-GAF¼ Modified Global Assessment of Functioning.
18 S. Kvrgic et al. / Psychiatry Research 209 (2013) 15–20
Table 2
Hierarchical regression analysis with STAR-P total score as dependent variable.
Variables to enter
Age
Duration of treatment by the current therapist
Positive Symptoms (PANSS)
Negative Symptoms (PANSS)
Depression (BDI-II)
Avoidant Attachment Style (PAM)
Recovery orientation (RAS)
Self-esteem decrement due to self-stigma (SSMIS)
Insight (BIS)
Adjusted R2
R2 change
Note: STAR-P ¼Scale to Assess Therapeutic Relationship-Patient Version; PANSS ¼ Positive and Negative Syndrome Scale; BDI-II ¼Beck
Depression Inventory; PAM¼Psychosis Attachment Measure; RAS ¼ Recovery Assessment Scale; Self-esteem decrement scale of the SSMIS;
BIS ¼Birchwood Insight Scale. R2 ¼ Proportion of explained variance.
n
po 0.05.
nn
p o 0.01.
The regression data are presented in Table 2. The analyses
revealed that the first block with the control variables explained a
significant amount (6%) of the STAR-P total score (adj. R2 ¼0.060,
F(6, 149) ¼2.87, p¼ 0.01). Within the first block, the PANSS
positive symptoms subscale contributed uniquely to the model’s
validity. The block with the variables recovery orientation and
self-stigma added significantly to validity with a further 12% of
explained variance of the STAR-P total score (adj. R2 ¼ 0.182, F(8,
147) ¼5.32, p ¼0.000). Within the second block, the RAS total
score and the SSMIS subscale Self-Esteem Decrement Due to Self-
Stigma contributed uniquely to validity, whereas PANSS positive
symptom was not significantly associated with therapeutic alli-
ance anymore. This change of significance points to a mediation
effect of recovery orientation and/or self-stigma between the
association of positive symptoms and therapeutic alliance (for
detailed explanation of mediation effects see Baron and Kenny,
1986). Finally, the addition of the third block with the BIS total
score to the equation yielded a significant validity increment of
4% (adj. R2 ¼0.220, F(9, 146)¼ 5.62, p¼0.000). In summary,
recovery orientation, insight and self-stigma explain about 16%
of the variance of the therapeutic alliance.
In order to elucidate whether recovery orientation or self-stigma
or both factors mediate the negative effect of positive symptoms on
therapeutic alliance, two separate linear regression analyses were
conducted. In both analyses, the independent variable was the
PANSS positive symptoms subscale. In the first analyses, we used
the RAS total score as dependent variable. There was no significant
relation between positive symptoms and RAS total score (adj.
R2 ¼0.00, F(1, 154)¼1.67, p¼ 0.20). In a second analysis we used
the SSMIS subscale Self-Esteem Decrement Due to Self-Stigma as
dependent variable. The analysis reveals a significant association
between positive symptoms and self-stigma (adj. R2 ¼0.076, F(1,
154)¼13.83, p¼0.00). Due to these results, it is suggested that the
negative impact of positive symptoms on therapeutic alliance may
be mediated by self-stigma.
4. Discussion
In a large sample of outpatients with schizophrenia or schizo-
affective disorder, a stronger recovery orientation, less self-stigma
and more insight contributed uniquely to a better therapeutic
alliance. Recovery orientation and self-stigma could explain a
total of 12% of the quality of therapeutic alliance, and insight
another 4%. A total of 22% of variance could be explained by our
regression model.
Step 1 (Beta) Step 2 (Beta) Step 3 (Beta)
0.013 0.035 0.046
0.106 0.058 0.042
 0.170n  0.150  0.146
0.034 0.061 0.067
 0.071  0.040  0.036
 0.114  0.026  0.039
0.369nn 0.393nn
 0.170n  0.150n
0.161n
0.060 0.182 0.220
0.122nn 0.038nn
In the maintenance phase of treatment, these associations were
stronger than relationships with positive and negative symptoms,
avoidant attachment style, age, and depression respectively. Also
the duration of treatment by the current therapist did not influence
the quality of the therapeutic alliance. This result might be
supported by the findings of Wittorf et al. (2010) which could
show that patients’ alliance ratings are relatively stable over the
course of treatments like CBT or supportive therapy.
The results of the present study further suggest that self-
stigma mediates the negative effect of positive symptoms on
therapeutic alliance. Positive symptoms have already been found
to be significantly related with self-stigma (Lysaker et al., 2007;
Yanos et al., 2008) and the results of the present study support
these findings. The mediation effect also might be a reason, why
the association between symptoms and patient’ therapeutic
alliance ratings are so far inconsistent. For example McCabe and
Priebe (2003) found hostility to explain 28% of the variance of
patient alliance ratings. In contrast, Couture et al. (2006) did not
report significant associations between clinical baseline charac-
teristics and patient-rated therapeutic alliance. Further studies
are needed for an additional clarification of mediating effects self-
stigma, positive symptoms and therapeutic alliance.
According to our hypothesis, insight into illness explained addi-
tional variance of the therapeutic alliance, although recovery orien-
tation and self-stigma were already integrated in the model.
Regarding that previous studies on this issue all used different
measures (Wittorf et al., 2009; Barrowclough et al., 2010; Lysaker
et al., 2011) and some of them also controlled for positive symptoms
(Wittorf et al., 2009) insight appears to be a moderately but
consistent predictor for therapeutic alliance when rated by patients.
Compared to other studies investigating outpatients with
schizophrenia, our sample demonstrates with 66% males compar-
able gender distribution (Rössler et al., 2005; Salize et al., 2009),
similar scores in role-functioning and symptom control (GAF;
Villalta-Gil et al., 2006; Rossi et al., 2009), psychotic symptoms
(PANSS; Chatterjee et al., 2003; Morrison et al., 2004), and
recovery orientation (RAS; Corrigan and Phelan, 2004) indicating
that our outpatient-sample could be regarded as representative
for people with schizophrenia treated in CMHC.
Our study has some methodological limitations. Results of the
regression analyses should be interpreted with caution in a causal
way because of the cross-sectional nature of the study design.
A longitudinal design is needed in order to test directionality of
the results. The non-significant associations of negative symp-
toms, avoidant attachment style, as well as depression could be
due to the restricted range of values whilst focusing a clinical
 S. Kvrgic et al. / Psychiatry Research 209 (2013) 15–20
more stable sample in maintenance phase of treatment. Further
studies e.g. with inpatients or younger individuals with a first
episode of schizophrenia are needed to establish the generality of
the results. Regarding that only 16% of the variance could be
explained by recovery orientation, self-stigma and insight, we
suggest that the assessed set of variables supposed to be asso-
ciated with therapeutic alliance may not be exhaustive, so
a multitude of other variables may be important for patients like
previous experiences with therapists, negative attitudes to med-
ications (Barrowclough et al., 2010) or factors related to the
therapist like empathy, expertness, attractiveness or trustworthi-
ness (Evans-Jones et al., 2009). Also the more or less high scores of
therapeutic alliance ratings may induce the problem of restricted
range of values so that associations between predicted variable
and predicting variables may underestimate the role of the
predicting variables. A further limitation is that some measures
were translated into German by the authors and not validated yet
(i.e. STAR-P; BIS; SSMIS; RAS). Validity and reliability of these
measures are not known and therefore the results of the present
study should be interpreted cautiously. Nevertheless, the Cron-
bach’s alphas of the measures in the present study were all
acceptable to good. Finally, future studies should also assess
quality of the therapeutic alliance from the therapist perspective,
as therapists and patients perspectives of alliance may not always
be convergent (Fitzpatrick et al., 2005; Wittorf et al., 2009).
Our findings have some important implications for clinical
practice in outpatient treatment. To build a powerful therapeutic
alliance, not only focusing insight by educational interventions
may be important, but also addressing recovery orientation and
self-stigma. Overcoming entrapment in illness and improving
recovery orientation in schizophrenia has already been outlined
in some more recent cognitive-behavioral approaches for psycho-
sis (CBTp; Shahar et al., 2004; Birchwood et al., 2005). So,
Andresen et al., (2003) suggested that the therapist should
support the patient in four recovery processes: finding hope; re-
establishment of identity; meaning in life; and responsibility for
recovery. These processes could be supported by assisting the
patient to develop more effective symptom management strate-
gies by CBTp (Vauth and Stieglitz, 2007; Lencer et al., 2011).
By doing this, the patient might strengthen his’ or hers’ self-
efficacy and promote recovery orientation. Self-stigma should be
addressed as goal of its own in CBT (Birchwood et al., 2007) by
cognitive challenging strategies, especially in the phases of treat-
ment where patients work out personal meaningful treatment
goals, and by this furthering empowerment of the patient (Vauth
et al., 2007; Kleim et al., 2008). Especially stigma-oriented
approaches like the cognitive-behavioral intervention of Knight
et al. (2006) might be applied. This approach coaches more
specifically functional coping skills, such as training how to advise
significant others about one’s illness in a meaningful and effective
way. Summarizing, CBTp may be a helpful method to work with
patients with schizophrenia, even if there are some mixed results
on its effect on outcome (see for an overview: Wykes et al., 2008;
Dickerson and Lehman, 2011).
In conclusion, low recovery orientation and increased self-stigma
might undermine the therapeutic alliance beyond the detrimental
effects of poor insight. This is an important, but seldom addressed
issue. Our findings might be of relevance for the improvement of
psychotherapeutic interventions for patients with schizophrenia.
Acknowledgments
Funding of this study was provided by the Swiss National
Science Foundation (SNSF, grant no.105314-120673).
19
References
Andresen, R., Oades, L., Caputi, P., 2003. The experience of recovery from
schizophrenia: towards an empirically validated stage model. Australian and
New Zealand Journal of Psychiatry 37, 586–594.
Baron, R.M., Kenny, D.A., 1986. The moderator-mediator variable distinction in
social psychological research: conceptual, strategic, and statistical considera-
tions. Journal of Personality and Social Psychology 51, 1173–1182.
Barrowclough, C., Meier, P., Beardmore, R., Emsley, R., 2010. Predicting therapeutic
alliance in clients with psychosis and substance misuse. Journal of Nervous
and Mental Disease 198, 373–377.
Beck, A.T., Steer, R.A., Brown, G.K., 1996. Manual for Beck Depression Inventory-II.
Psychological Corporation, San Antonio.
Bentall, R., Tarrier, N., Lewis, S., Haddock, G., Kinderman, P., Kingdon, D., 2002. the
SoCRATES Team, 2002. The therapeutic alliance in early psychosis. Acta
Psychiatrica Scandinavica (Suppl. 106), S94.
Berry, K., Barrowclough, C., Wearden, A., 2008. Attachment theory: a framework
for understanding symptoms and interpersonal relationships in psychosis.
Behaviour Research and Therapy 46, 1275–1282.
Berry, K., Wearden, A., Barrowclough, C., Liversidge, T., 2006. Attachment styles,
interpersonal relationships and psychotic phenomena in a non-clinical student
sample. Personality and Individual Differences 41, 707–718.
Birchwood, M., Iqbal, Z., Upthegrove, R., 2005. Psychological pathways to depres-
sion in schizophrenia: studies in acute psychosis, post psychotic depression
and auditory hallucinations. European Archives of Psychiatry and Clinical
Neuroscience 255, 202–212.
Birchwood, M., Smith, J., Drury, V., Healy, J., Macmillan, F., Slade, M., 1994. A self-
report insight scale for psychosis: reliability, validity and sensitivity to change.
Acta Psychiatrica Scandinavica 89, 62–67.
Birchwood, M., Trower, P., Brunet, K., Gilbert, P., Iqbal, Z., Jackson, C., 2007. Social
anxiety and the shame of psychosis: a study in first episode psychosis.
Behaviour Research and Therapy 45, 1025–1037.
Bordin, E.S., 1979. The generalizability of the psychoanalytic concept of the
working alliance. Psychotherapy: Theory Research and Practice 16, 252–260.
Catty, J., Winfield, H., Clement, S., 2007. The therapeutic relationship in secondary
mental health care: a conceptual review of measures. Acta Psychiatrica
Scandinavica 116, 238–252.
Cavelti, M., Kvrgic, S., Beck, E.M., Kossowsky, J., Vauth, R., 2011. Assessing recovery
from schizophrenia as an individual process. A review of self-report instru-
ments. European Psychiatry 27, 19–32.
Cavelti, M., Beck., E.-M., Kvrgic, S., Kossowsky, J., Vauth, R., 2012. The role of
subjective illness beliefs and attitude toward recovery within the relationship
of insight and depressive symptoms among people with schizophrenia
spectrum disorders. Journal of Clinical Psychology 68, 462–476.
Chatterjee, S., Patel, V., Chatterjee, A., Weiss, H.A., 2003. Evaluation of a
community-based rehabilitation model for chronic schizophrenia in rural
India. British Journal of Psychiatry 182, 57–62.
Chiu, M.Y., Ho, W.W., Lo, W.T., Yiu, M.G., 2009. Operationalization of the SAMHSA
model of recovery: a quality of life perspective. Quality of Life Research 19, 1–13.
Corrigan, P., Markowitz, F., Watson, A., 2004a. Structural levels of mental illness
stigma and discrimination. Schizophrenia Bulletin 30, 481–491.
Corrigan, P.W., Giffort, D., Rashid, F., Leary, M., Okeke, I., 1999. Recovery as a
psychological construct. Community Mental Health Journal 35, 231–239.
Corrigan, P.W., Larson, J.E., Rusch, N., 2009. Self-stigma and the ‘‘why try’’ effect:
impact on life goals and evidence-based practices. World Psychiatry 8, 75–81.
Corrigan, P.W., Phelan, S.M., 2004. Social support and recovery in people with
serious mental illnesses. Community Mental Health Journal 40, 513–523.
Corrigan, P.W., Salzer, M., Ralph, R.O., Sangster, Y., Keck, L., 2004b. Examining the
factor structure of the recovery assessment scale. Schizophrenia Bulletin 30,
1035–1041.
Corrigan, P.W., Watson, A.C., 2002. Understanding the impact of stigma on people
with mental illness. World Psychiatry 1, 16–20.
Corrigan, P.W., Watson, P.W., Barr, L., 2006. The self-stigma of mental illness:
implications for self-esteem and self-efficacy. Journal of Social and Clinical
Psychology 25, 875–884.
Couture, S.M., Roberts, D.L., Penn, D.L., Cather, C., Otto, M.W., Goff, D., 2006. Do
baseline client characteristics predict the therapeutic alliance in the treatment
of schizophrenia? Journal of Nervous and Mental Disease 194, 10–14.
Dickerson, F.B., Lehman, A.F., 2011. Evidence-based psychotherapy for schizo-
phrenia: 2011 update. Journal of Nervous and Mental Disease 199, 520–526.
Dozier, M., Lomax, L., Tyrrell, C.L., Lee, S.W., 2001. The challenge of treatment for
clients with dismissing states of mind. Attachment and Human Development
3, 62–76.
Draine, J., Solomon, P., 1996. Case manager alliance with clients in an older cohort.
Community Mental Health Journal 32, 125–134.
Drake, R.J., Pickels, A., Bentall, R.P., Kinderman, P., Haddock, G., Tarrier, N., Lewis,
S.W., 2004. The evolution of insight, paranoia and depression during early
schozophrenia. Psychological Medicine 34, 285–292.
Dunn, H., Morrison, A.P., Bentall., R.P., 2006. The relationship between patient
suitability, therapeutic alliance, homework compliance and outcome in
cognitive therapy for psychosis. Clinical Psychology and Psychotherapy 13,
145–152.
Evans-Jones, C., Peters, E., Barker, C., 2009. The therapeutic relationship in CBT for
psychosis: client, therapist and therapy factors. Behavioural and Cognitive
Psychotherapy 37, 527–540.
20 S. Kvrgic et al. / Psychiatry Research 209 (2013) 15–20
Fitzpatrick, M.R., Iwakabe, S., Stalikas, A., 2005. Perspective divergence in the
working alliance. Psychotherapy Research 15, 69–80.
Francis, J.L., Penn, D.L., 2001. The relationship between insight and social skill in
persons with severe mental illness. Journal of Nervous and Mental Disease
189, 822–829.
Frank, A.F., Gunderson, J.G., 1990. The role of the therapeutic alliance in the
treatment of schizophrenia. Relationship to course and outcome. Archives of
General Psychiatry 47, 228–236.
Gehrs, M., Goering, P., 1994. The relationship between the working alliance and
rehabilitation outcomes of schizophrenia. Psychosocial Rehabilitation Journal
18, 43–54.
Gibbons, M.B.C., Crits-Christoph, P., de la Cruz, C., Barber, J.P., Siqueland, L., Gladis,
M., 2003. Pretreatment expectations, interpersonal functioning and symptoms
in the prediction of the therapeutic alliance across supportiveexpressive
psychotherapy and cognitive therapy. Psychotherapy Research 13 (1), 59–76.
Hall, R.C., 1995. Global assessment of functioning. A modified scale. Psychoso-
matics 36, 267–275.
Hasson-Ohayon, I., Kravetz, S., Meir, T., Rozencwaig, S., 2009. Insight into severe
mental illness, hope, and quality of life of persons with schizphrenia and
schizoaffective disorders. Psychiatry Research 167, 231–238.
Horvath, A.O., Symonds, B.D., 1991. Relation between working allinace and
outcome in psychotherapy: a meta-analysis. Journal of Counseling Psychology
38, 139–149.
Johnson, D.P., Penn, D.L., Bauer, D.J., Meyer, P., Evans, E., 2008. Predictors of the
therapeutic alliance in group therapy for individuals with treatment-resistant
auditory hallucinations. British Journal of Clinical Psychology 47, 171–183.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale
(PANSS) for schizophrenia. Schizophrenia Bulletin 13, 261–276.
Kleim, B., Vauth, R., Adam, G., Stieglitz, R.D., Hayward, P., Corrigan, P., 2008.
Perceived stigma predicts low self-efficacy and poor coping in schizophrenia.
Journal of Mental Health 17, 482–491.
Klinkenberg, W.D., Calsyn, R.J., Morse, G.A., 1998. The helping alliance in case
management for homeless persons with severe mental illness. Community
Mental Health Journal 34, 569–578.
Knight, M.T.D., Wykes, T., Hayward, P., 2006. Group treatment of perceived stigma
and self-esteem in schizophrenia: a waiting list trial of efficacy. Behavioural
and Cognitive Psychotherapy 34, 305–318.
Kvrgic, S., Beck, E.M., Cavelti, M., Kossowsky, J., Stieglitz, R.D., Vauth, R., 2011.
Focussing on the adult attachment style in schizophrenia in Community
Mental Health Centres: validation of the Psychosis Attachment Measure
(PAM) in a German-speaking sample. International Journal of Social Psychiatry
58, 362–373.
Lecomte, T., Spidel, A., Leclerc, C., MacEwan, G.W., Greaves, C., Bentall, R.P., 2008.
Predictors and profiles of treatment non-adherence and engagement in
services problems in early psychosis. Schizophrenia Research 102, 295–302.
Lencer, R., Harris, M.S.H., Weiden, P.J., Stieglitz, R.D., Vauth, R., 2011. When
Psychopharmacology is Not Enough. Using Cognitive Behavioral Therapy
Techniques for Persons with Persistent Psychosis. Hogrefe, Göttingen.
Lincoln, T.M., Lullmann, E., Rief, W., 2007. Correlates and long-term consequences
of poor insight in patients with schizphrenia. A systematic review. Schizo-
phrenia Bulletin 33, 1342–2324.
Livingston, J.D., Boyd, J.E., 2010. Correlates and consequences of internalized
stigma for people living with mental illness: a systematic review and meta-
analysis. Social Science and Medicine 71, 2150–2161.
Lysaker, P.H., Bell, M.D., Bryson, G.J., Kaplan, E.Z., 1998. Insight and interpersonal
function in schizophrenia. Journal of Nervous and Mental Disease 186,
432–436.
Lysaker, P.H., Davis, L.W., Buck, K.D., Outcalt, S., Ringer, J.M., 2011. Negative
symptoms and poor insight as predictors of the similarity between client and
therapist ratings of therapeutic alliance in cognitive behavior therapy for
patients with schizophrenia. Journal of Nervous and Mental Disease 199,
191–195.
Lysaker, P.H., Roe, D., Yanos, P.T., 2007. Toward understanding the insight paradox:
internalized stigma moderates the assoctiation between insight and social
functioning, hope, and self-esteem among people with schizophrenia spec-
trum disorders. Schizophrenia Bulletin 33, 192–199.
Martin, D.J., Garske, J.P., Davis, M.K., 2000. Relation of the therapeutic alliance with
outcome and other variables: a meta-analytic review. Journal of Consulting
and Clinical Psychology 68, 438–450.
McCabe, R., Priebe, S., 2003. Are therapeutic relationships in psychiatry explained
by patients’ symptoms? Factors influencing patient ratings. European Psy-
chiatry 18, 220–225.
McGuire-Snieckus, R., McCabe, R., Catty, J., Hansson, L., Priebe, S., 2007. A new
scale to assess the therapeutic relationship in community mental health care:
STAR. Psychological Medicine 37, 85–95.
Mintz, A.R., Dobson, K.S., Romney, D.M., 2003. Insight in schizophrenia: a meta-
analysis. Schizophrenia Research 61, 75–88.
Mohamed, S., Rosenheck, R., McEvoy, J., Swartz, M., Stroup, S., Lieberman, J.A.,
2009. Cross-sectional and longitudinal relationships between insight and
attitudes toward medication and clinical outcomes in chronic schizophrenia.
Schizophrenia Bulletin 35, 336–346.
Morrison, A.P., Renton, J.C., Williams, S., Dunn, H., Knight, A., Kreutz, M., Nothard,
S., Patel, U., Dunn, G., 2004. Delivering cognitive therapy to people with
psychosis in a community mental health setting: an effectiveness study. Acta
Psychiatrica Scandinavica 110, 36–44.
Restifo, K., Harkavy-Friedman, J.M., Shrout, P.E., 2009. Suicidal behaviour in
schizophrenia: a test of the demoralization hypothesis. Journal of Nervous
and Mental Disease 197, 147–153.
Rossi, A., Pacifico, R., Stratta, P., 2009. Attitudes toward medication and the clinical
variables in schizophrenia: structural equation models. Journal of Patient
Preference and Adherence 3, 305–309.
Rössler, W., Salize, H.J., van Os, J., Riecher-Rossler, A., 2005. Size of burden of
schizophrenia and psychotic disorders. European Neuropsychopharmacology
15, 399–409.
Rüsch, N., Corrigan, P.W., Wassel, A., Michaels, P., Larson, J.E., Olschewski, M.,
Wilkniss, S., Batia, K., 2009. Self-stigma, group identification, perceived
legitimacy of discrimination and mental health service use. British Journal of
Psychiatry 195, 551–552.
Rüsch, N., Holzer, A., Hermann, C., Schramm, E., Jacob, G.A., Bohus, M., Lieb, K.,
Corrigan, P.W., 2006. Self-stigma in women with borderline personality
disorder and women with social phobia. Journal of Nervous and Mental
Disease 194, 766–773.
Salize, H.J., McCabe, R., Bullenkamp, J., Hansson, L., Lauber, C., Martinez-Leal, R.,
Reinhard, I., Rossler, W., Svensson, B., Torres-Gonzalez, F., van den Brink, R.,
Wiersma, D., Priebe, S., 2009. Cost of treatment of schizophrenia in six
European countries. Schizophrenia Research 111, 70–77.
Shahar, G., Trower, P., Iqbal, Z., Birchwood, M., Davidson, L., Chadwick, P., 2004.
The person in recovery from acute and severe psychosis: the role of
dependency, self-criticism, and efficacy. American Journal of Orthopsychiatry
74, 480–488.
Shrout, P.E., Fleiss, J.L., 1979. Interclass correlations: uses in assessing rater
reliability. Psychological Bulletin 86, 420–428.
Staring, A.B., Van der Gaag, M., Van den Berge, M., Duivenvoorden, H.J., Mulder,
C.L., 2009. Stigma moderates the associations of insight with depressed mood,
low self-esteem, and low quality of life in patients with schizophrenia
spectrum disorders. Schizophrenia Research 115, 363–369.
Stieglitz, R.D., Fähndrich, E., Möller, H.J., 1998. Syndromale Diagnostik psychischer
Störungen. Hogrefe, Göttingen.
Svensson, B., Hansson, L., 1999. Relationships among patient and therapist ratings
of therapeutic alliance and patient assessments of therapeutic process: a study
of cognitive therapy with long-term mentally ill patients. Journal of Nervous
and Mental Disease 187, 579–585.
Vauth, R., Kleim, B., Wirtz, M., Corrigan, P.W., 2007. Self-efficacy and empower-
ment as outcomes of self-stigmatizing and coping in schizophrenia. Psychiatry
Research 150, 71–80.
Vauth, R., Stieglitz, R.D., 2007. Chronisches Stimmenhören und persistierender
Wahn. Fortschritte der Psychotherapie. Hogrefe, Göttingen.
Villalta-Gil, V., Vilaplana, M., Ochoa, S., Haro, J.M., Dolz, M., Usall, J., Cervilla, J.,
2006. Neurocognitive performance and negative symptoms: are they equal in
explaining disability in schizophrenia outpatients? Schizophrenia Research 87,
246–253.
Vogel, D.L., Wade, N.G., Haake, S., 2006. Measuring self-stigma associated with
seeking psychological help. Journal of Counseling Psychology 53, 325–337.
Waldheter, E.J., Penn, D.L., Perkins, D.O., Mueser, K.T., Whaley Owens, L., Cook, E.,
2008. The Graduated Recovery Intervention Programm for first episode
psychosis: Treatment development and preliminary data. Community Mental
Health Journal 44, 443–455.
Webb, C.A., DeRubeis, R.J., Amsterdam, J.D., Shelton, R.C., Hollon, S.D., Dimidjian, S.,
2011. Two aspects of the therapeutic alliance: differential relations with
depressive symptom change. Journal of Consulting and Clinical Psychology
79, 279–283.
Wittchen, H.-U., Wunderlich, U., Gruschwitz, S., Zaudig, M., 1997. Strukturiertes
Klinisches Interview für DSM-IV, 1st ed. Hogrefe, Göttingen.
Wittorf, A., Jakobi, U., Bechdolf, A., Muller, B., Sartory, G., Wagner, M., Wiedemann,
G., Wolwer, W., Herrlich, J., Buchkremer, G., Klingberg, S., 2009. The influence
of baseline symptoms and insight on the therapeutic alliance early in the
treatment of schizophrenia. European Psychiatry 24, 259–267.
Wittorf, A., Jakobi, U.E., Bannert, K.K., Bechdolf, A., Müller, B.W., Sartory, G.,
Wagner, M., Wiedemann, G., Wölwer, W., Herrlich, J., Buchkremer, G.,
Klingberg, S., 2010. Does the cognitive dispute of psychotic symptoms do
harm to the therapeutic alliance? Journal of Nervous and Mental Disease 198,
478–485.
Wykes, T., Steel, C., Everitt, B., Tarrier, N., 2008. Cognitive behavior therapy for
schizphrenia: effect sizes, clinical models, and methodological rigor. Schizo-
phrenia Bulletin 34, 523–537.
Yanos, P.T., Roe, D., Markus, K., Lysaker, P.H., 2008. Pathways between internalized
stigma and outcomes related to recovery in schizophrenia spectrum disorders.
Psychiatric Services 59, 1437–1442.