Psychiatry Research 257 (2017) 315–321

Contents lists available at ScienceDirect

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

Medication adherence in schizophrenia: The role of insight, therapeutic MARK

alliance and perceived trauma associated with psychiatric care
⁎
Arnaud Tessiera,b, , Laurent Boyerc, Mathilde Huskyd, Franck Baylée, Pierre-Michel Llorcaf,
David Misdrahia,b
a
INCIA-CNRS UMR 5287, Université Bordeaux, 33076 Bordeaux, France
b
Pole de Psychiatrie Adulte et Universitaire, C.H. Charles Perrens, 33076 Bordeaux cedex, France
c
Université de Aix-Marseille, EA 3279–Santé Publique, Maladies Chroniques et Qualité de Vie–Unité de Recherche, France
d
Université de Bordeaux, Laboratoire de Psychologie EA4149, Institut Universitaire de France, 33000 Bordeaux, France
e
Université Paris V, Service Hospitalo-Universitaire de Santé Mentale et Thérapeutique, Hôpital Sainte-Anne, 75014 Paris, France
f
CHU de Clermont-Ferrand, Service de Psychiatrie, Place Henri Dunant, 63000 Clermont-Ferrand, France

A R T I C L E I N F O A B S T R A C T

Keywords: Medication non adherence in schizophrenia is a major cause of relapse and hospitalization and remains for
Schizophrenia clinicians an important challenge. This study investigates the associations between insight, therapeutic alliance,
Medication adherence perceived trauma related to psychiatric treatment and medication adherence in patients with schizophrenia. In
Coercion this multicenter study, 72 patients were assessed regarding symptomatology, self-reported adherence with
Therapeutic alliance
medication, insight, medication side-effects, therapeutic alliance and perceived trauma related to psychiatric
Insight
treatment. Structural Equation Modeling (SEM) was used to test predicted paths among these variables. The data
fit a model in which medication adherence was directly predicted by insight, therapeutic alliance and perceived
trauma related to psychiatric treatment. Perceived trauma moderates the role of insight on medication ad-
herence. The final model showed good fit, based on four reliable indices. Greater adherence was correlated with
higher insight, higher therapeutic alliance and lower perceived trauma. These three variables appear to be
important determinants of patient's medication adherence. Medication adherence could be enhanced by redu-
cing perceived trauma and by increasing insight. The need for mental health providers to acknowledge patients'
potentially traumatic experience with psychiatric treatment and the need to encourage greater involvement in
care are discussed.

1. Introduction factors including attitudes toward medication (Baloush-Kleinman et al.,

Rates of medication non-adherence among patients with schizo-
phrenia have been estimated at approximately 50% (Gilmer et al.,
2004; Lacro et al., 2002; Velligan et al., 2009), leading to higher rates of
relapse and hospitalization as well as to decreasing clinical, cognitive
and functional prognosis (Ascher-Svanum et al., 2006; Llorca, 2008;
Robinson et al., 1999; Weiden et al., 2004). While the identification of
determinants of poor adherence has yielded valuable results, additional
research is needed as the identification and characterization of cluster
of patients with poor medication adherence remains an important
challenge (Misdrahi et al., 2016).
Insight into illness is commonly observed as a determinant of poor
adherence (Mohamed et al., 2009; Novick et al., 2015; Rocca et al.,
2008) but its effect may be indirect and mediated by other potential
2011; Beck et al., 2011; Drake et al., 2015; Samalin et al., 2016),
therapeutic alliance (Day et al., 2005; Roche et al., 2014) or perceived
coercion (Day et al., 2005; Jaeger and Rossler, 2010). Engaging patients
in treatment requires an efficient therapeutic alliance (Frank and
Gunderson, 1990) which is a key factor in the care of patients with
severe psychiatric disorders as it is associated with better adherence
(Lecomte et al., 2008; McCabe et al., 2012; Misdrahi et al., 2012).
Despite the hypothesized association between the therapeutic re-
lationship and the experience of coercion (Gilburt et al., 2008), litera-
ture is scarce to test quantitatively this link. Few studies have in-
vestigated whether traumatic and coercive experiences related to
psychiatric care are associated with medication adherence.
That being said, most studies have examined the impact of each
determinant in isolation. However, from a methodological perspective,

⁎
Correspondence to: CH Charles Perrens, 121 rue de la Béchade, CS81285, 33076 Bordeaux cedex, France.
E-mail addresses: atessier@ch-perrens.fr (A. Tessier), laurent.boyer@ap-hm.fr (L. Boyer), mathilde.husky@u-bordeaux.fr (M. Husky), f.bayle@ch-sainte-anne.fr (F. Baylé),
pmllorca@chu-clermontferrand.fr (P.-M. Llorca), david.misdrahi@u-bordeaux.fr (D. Misdrahi).

http://dx.doi.org/10.1016/j.psychres.2017.07.063
Received 8 February 2017; Received in revised form 30 May 2017; Accepted 29 July 2017
Available online 31 July 2017
0165-1781/ © 2017 Elsevier B.V. All rights reserved.
A. Tessier et al.
these studies did not use analytical techniques that allow for an ex-
amination of the direction and the structure of the relations between
the determinants and medication adherence. More specifically, these
studies did not differentiate direct and indirect determinants of ad-
herence. Some of the determinants that have no direct effect may have
an indirect effect through mediating factors.
The aim of the present study was to investigate the complex re-
lationship among insight, therapeutic alliance, perceptions of trauma
experiences related to psychiatric treatment and medication adherence
in patients with schizophrenia. We used structural equation modeling
(SEM), which is a useful statistical procedure, to test a theory involving
non-straightforward relationships and is therefore well suited to the
management of cross-sectional data for inferential purposes.
2. Methods
2.1. Participants
In this multicenter study, 72 patients were recruited consecutively
in three comparable psychiatric hospitals located in Bordeaux (n = 47),
Clermont-Ferrand (n = 8) and Paris (n = 17). Recruited patients did
not differ on socio-demographic data between the three centers.
Patients were recruited during hospitalization, less than one week prior
to scheduled discharge, after the remission of acute symptomatology,
when patients were considered to be clinically stable (judged compa-
tible with answering questionnaires) and ambulatory routine care was
organized. Once discharged, patients received post-discharge care as
usual with a monthly appointment with their psychiatrist. Inclusion
criteria were (1) a diagnosis of schizophrenia or schizoaffective disorder
according to DSM-IV-TR criteria (American Psychiatric Association,
2000), (2) at least 18 years old, (3) able to understand the protocol, and
(4) fluent French speaker. Exclusion criteria included traumatic head
injury, any past or present major medical or neurological illness and
mental retardation. The study conformed to France's laws on bioethics
and clinical research, and on data protection (CPP-Ile de France III, N°
ID RCB: 2008-A00504-51). All subjects gave their written informed
consent to participate in this study.
2.2. Measures
Socio-demographic data were collected using a standardized semi-
structured questionnaire. Variables included the following: age, gender,
marital status, parental status, educational level and employment
status.
The severity of psychopathology was assessed by the PANSS (Kay
et al., 1987), which comprises three different subscales (positive, ne-
gative and general psychopathology). Current and lifetime comorbid
substance use disorders were determined using the MINI International
Neuropsychiatric Interview (Lecrubier et al., 1997). Overall functioning
was estimated with the Global Assessment of Functioning scale (GAF)
(American Psychiatric Association, 2000). Depressive symptoms were
assessed using the Beck Depression Inventory (BDI) (Beck et al., 1988).
Perceived traumatic experiences associated with prior psychiatric
treatment and hospitalization was indicated by a positive response to
the following question: “Have you ever been the witness or the subject
of events perceived as traumatic during your care or hospitalization
(s)?”. To specify the context of subjective traumatic experiences the
following items were explored: “receiving forced medication like in-
jection drugs, physical restraint by belt or restraint by security ties or
being put in seclusion”. The responses to the latter items were not in-
cluded in the analyses.
Insight into illness was assessed using the Birchwood Insight Scale
(BIS) (Birchwood et al., 1994). This is an 8-item scale that has been
shown to have good reliability and validity in people who experience
acute and chronic psychotic symptoms. The BIS scored on a three-point
Likert-type scale (from 0 to 2). The total score for the BIS ranges from 0
316
Psychiatry Research 257 (2017) 315–321
to 12. Higher scores indicate greater insight.
Medication Adherence was evaluated using the French translation
of the MARS (Fond et al., 2016; Misdrahi et al., 2016). The sum of items
yields a total score ranging from 0 (poor adherence to treatment) to 10
(good adherence to treatment). Patients were asked to report on their
medication adherence over the 7 days preceding hospitalization.
Therapeutic alliance with the prescriber was obtained on a self-re-
ported 4-Point ordinal Alliance Scale (4PAS) (Misdrahi et al., 2009). A
higher 4PAS score indicates greater therapeutic alliance.
Side-effects using the UKU (Udvalg for Kliniske Undersogelser) side-
effect rating scale (Lingjaerde et al., 1987).
2.3. Statistical analysis
Descriptive data are first presented: quantitative parameters are
described with means and standard deviations, qualitative parameters
in frequencies and percentages. Normality of quantitative data was
checked using the Kolmogorov–Smirnov 1-sample test. The pairwise
covariance and Pearson correlation matrices for measured variables and
medication adherence were presented. All tests were two-sided, and
statistical significance was defined as P < 0.01 to correct for multiple
hypothesis testing.
Then, Structural Equation Model (SEM) (Davies et al., 2016;
Gunzler et al., 2013; Thomas et al., 2017) was used to examine the
hypothetical relationships between variables. The final model was ob-
tained by a step-by-step procedure (Muthén and Muthén, 1998), with
strong indices evaluating the model (chi-square, RMSEA, CFI and SRMR
indices) and did not include GAF, BDI, comorbidity and demographic.
In the end, the 5 latent variables include were insight (BIS), sympto-
matology (PANSS total), therapeutic alliance (4PAS), adherence
(MARS) and perceived trauma related to psychiatric treatment. PANSS
total score was used to prevent collinearity abnormality.
The level of significance was set at p < 0.05. Model fit was eval-
uated using four indices:
• Chi-square: a value less than 3 indicates that the observed correla-
tions are not significantly different than the expected correlations;
• RMSEA (Root Mean Square Error of Approximation): indicates how
well the model would fit the hypothetical population covariance
matrix. A value lower than 0.05 is indicative of a close-fitting model,
between 0.05 and 0.08 a reasonable fit, and 0.10 or greater a poor
model;
• CFI (Comparative Fit Index): indicates the extent to which the model
provides a better fit than the null model, range from 0 to 1. A value
Greater than 0.90 suggests a good fit;
• SRMR (Standardized Root Mean square Residuals): the average
difference between the correlations predicted by the model and the
observed correlations. A value less than 0.10 indicate a good fit.
On the SEM model, links are modeled by paths with a coefficient,
which can range from −1 to +1, indicating the strength and direction
of the paths. The significance of the path coefficient is assessed using
the standard errors and t-values for each coefficient. In addition to the
statistical significance of the paths coefficients, the strength of the re-
lationship plays a role in determining whether the relationships are
weak (< 0.2), moderate (0.2–0.5) or strong (> 0.5).
Analyses were performed with the Statistical Procedures for Social
Sciences (SPSS v.20) (SPSS Inc, an IBM Company, Chicago, IL) and
Structural Equation Modeling was carried out using MPlus Version 7.1
software for Windows (Muthén and Muthén, 1998).
3. Results
3.1. Sample characteristics (Table 1)
Of the 72 participants, the majority met criteria for schizophrenia
A. Tessier et al. Psychiatry Research 257 (2017) 315–321

Table 1 lower perceived trauma. Among clinical variables, PANSS scores, side-
Clinical and demographic characteristics of study sample (N = 72). effects (UKU) and global functioning scores did not significantly cor-
relate with medication adherence Table 3.
Variables N %

Sex, male 49 68.1 3.3. Structural equation model

Diagnostic, schizoaffective disorder 22 30.6
Current substance, use or abuse 16 22.2
The structural equation model fitted to assess the hypothesized
Antipsychotic, SGA 70 97.2
Perceived trauma, yes 40 55.6 model is illustrated in Fig. 1. Three paths were not significant and re-
Lifetime suicide attempt, yes 34 47.2 moved from the theoretical model: insight–alliance, perceived trau-
Marital Status, never married, separated or divorced 67 93.1 ma–alliance and symptomatology (PANSS)–perceived trauma. The final
Children, yes 8 11.1 model showed good fit based on the chi-square statistic (normed χ2 =
Educational level, primary or lower 45 62.5
0.49), RMSEA = 0.06, CFI = 1.00, and SRMR = 0.04. The SEM re-
Employment status, employed 18 25.0
Mean SD vealed two different types of interactions: direct effects and indirect
Age (years) 38.7 11.5 effects. Five direct significant but moderate associations were found
Age at first hospitalization (years) 29.1 11.3 between (1) insight on perceived trauma (path coefficient = −0.27),
Length of illness (years) 12.6 7.4
(2) perceived trauma on medication adherence (path coefficient =
Number of previous hospitalizations (N) 6.1 5.5
−0.25), (3) insight on medication adherence (path coefficient = 0.22),
Mean (SD): mean (standard deviation). (4) symptomatology on alliance (path coefficient = −0.46) and (5)
BMI: Body Mass Index. alliance on medication adherence (path coefficient = 0.35). Indirect
SGA: Second Generation of Antipsychotic. effects were found between (1) insight on medication adherence via
perceived trauma (path coefficient = 0.22), (2) symptomatology on
(69.4%), and the remaining 30.6% met criteria for schizoaffective dis- medication adherence via alliance (path coefficient = −0.17).
order. The mean age was 38.7 (SD = 11.5) years, and 49 patients
(68.1%) were men. Overall, 27 patients (37.5%) had a university level 4. Discussion
education. The mean length of illness was 12.6 years (SD = 7.4 years)
and the mean number of previous hospitalizations was 6.1 (SD = 5.5). Overall, the present study revealed that in a sample of stabilized
Psychometrics measures revealed that patients had a high level of schizophrenic patients, lower adherence to pharmacological treatments
psychotic symptoms with a total PANSS score of 89.9 (SD = 19.2) was significantly associated with lower insight, lower therapeutic alli-
(Table 2). Therapeutic alliance and insight were generally high (mean ance and perceived treatment-related trauma. These associations were
35.7, S.D. 7.4; mean 8.03, S.D. 3.12 respectively). Medication ad- then explored using structural equation modeling. The model showed
herence was moderate with an average score of 5.6 (SD = 1.9). Side- good fit with the data and provided three main findings: (1) therapeutic
effect and depressive symptoms ratings were low. Patients’ ratings in- alliance was found to be a strong predictor of medication adherence, (2)
dicated that 55.6% reported treatment-related traumatic experiences. to a lesser degree, insight was linked to medication adherence whereas
treatment-related trauma moderates weakly the role of insight on
3.2. Correlations between variables (Table 3) medication adherence, and (3) therapeutic alliance moderates the role
of psychotic symptoms on medication adherence.
Pearson's correlations revealed strong correlations between the sub- In accordance with previous cross sectional studies, insight (Amador
scores and the total score of the PANSS, as expected. Pearson's corre- et al., 1994; Lacro et al., 2002; Lincoln et al., 2007) and therapeutic
lation between medication adherence, and perceived trauma, insight alliance (Day et al., 2005; Frank and Gunderson, 1990; Misdrahi et al.,
and therapeutic alliance were significant (r = −0.316, r = 0.374 and r 2012; Novick et al., 2015) were found to be two strong predictors of
= 0.413, respectively). The correlation between medication adherence, medication adherence. These results were confirmed in a longitudinal
therapeutic alliance (r = 0.413), perceived trauma (r = −0.316) and study in a sample of 112 participants with schizophrenia or schi-
insight (r = 0.374) supporting the assumption that higher adherence zoaffective disorder followed for 6 months (Baloush-Kleinman et al.,
rating was related to higher insight, higher therapeutic alliance and 2011). Adherent participants assessed with a Visual Analog Scale and
rated by participants, relatives, and treating physicians showed more
Table 2 insight into their illness, awareness of the need for medication and
Summary data for variables included in the study. positive perceptions of trust in the patient–physician therapeutic alli-
ance. Structural equation modeling also showed that over 6 months,
Variables No. of patients Mean SD Range
symptom severity as assessed by the CGI predicted adherence. In con-
MARS 71 5.63 1.86 1–9 trast with the latter study, the present findings indicated that symptom
BIS 58 8.03 3.12 2–12 severity using the PANSS was indirectly related to medication ad-
4PAS 70 35.69 7.44 11–44 herence. These contrasting findings might be explained by the use of a
BDI 70 7.29 6.31 0–39
different instrument to assess symptom severity and adherence. More-
PANSS–Positive 65 21.25 6.29 11–35
PANSS–Negative 65 21.42 6.40 9–39 over, the population in the present study was older, with a more severe
PANSS–General 65 47.28 10.97 22–72 illness when comparing GAF scores and number of prior hospitaliza-
PANSS Total 65 89.94 19.20 47–130 tions. Indeed, a total mean score for the PANSS of 89.9 is usually in-
GAF 66 45.59 12.38 21–70 terpreted as reflecting moderate psychotic symptomatology. Given the
UKU 64 7.58 5.84 0–22
association between insight, therapeutic alliance and medication ad-
Abbreviations:. herence, specific approaches might be useful to improve therapeutic
Mean (SD): mean (standard deviation). alliance with the treating team in patients with low insight, as this al-
MARS: Medication Adherence Rating Scale. liance impacts directly on adherence.
BIS: Birchwood Insight Scale. In contrast with prior research, insight was not directly associated
4PAS: 4 Point Alliance Scale.
with therapeutic alliance in the present study. One study (Wittorf et al.,
BDI: Beck Depression Inventory.
PANSS: Positive and Negative Syndrome Scale. 2009) demonstrated that a lack of insight affects the therapeutic alli-
GAF: Global Assessment of Functioning. ance as perceived by the patient. Discrepancy with our results could be
UKU: Udvalg for Kliniske Undersogelser Side Effects Rating Scale. explained by the different study design. In the latter study, therapeutic

317
A. Tessier et al.
alliance was explored in an ongoing multicentric, randomized con-
trolled clinical trial comparing a cognitive-behavioral intervention to
supportive therapy for patients with persistent positive symptoms.
Therefore, the therapeutic alliance scale was used to assess efficacy of a
psychotherapeutic intervention rather than in the current study. The
4PAS scale assessed the therapeutic alliance with the psychiatrist in the
context of routine mental health. Finally, as explained by the authors,
symptoms and insight together explained not more than 10% of the
variance of the therapeutic alliance suggesting that other factors not yet
determined explain the greater part of the alliance variance.
More than half of the participants reported a prior treatment-related
traumatic experience. In addition, our data confirmed the direct influ-
ence of perceived trauma on adherence, which could also moderate the
role of insight on medication adherence. This result is consistent with
prior a study (Paksarian et al., 2014) in which 69% of participants re-
ported that at least one of their hospitalizations was traumatic or ex-
tremely distressing in a 10-year prospective cohort of patients with
psychosis hospitalized between 1989 and 1995. As in the current study,
perceived trauma associated with psychiatric hospitalization was as-
sessed by an interviewer-administered questionnaire and self-reported
forced medication was significantly associated with less time spent in
treatment over the ten-year period of this prospective cohort. The
current findings are consistent with prior research suggesting that pa-
tients who felt coerced at hospital admission were less likely to take
medications or use mental health services than were those who did not
feel coerced (Kaltiala-Heino et al., 1997). Attitudes towards psychiatric
coercive interventions in healthy individuals and patients with schizo-
phrenia, were investigated by a six case vignette depicting scenarios of
ethical dilemmas and demanding decisions in favor of or against
coercive interventions (Mielau et al., 2015). In accordance with the
evidence substantiating the role of insight into psychotic disorders and
its relevance for treatment adherence, regression analyses confirmed
that patients displaying higher insight scores frequently viewed the use
of coercive interventions as warranted. This was most prominent in
patients who had previously been mechanically restrained. As sug-
gested by Jaeger et al. (Jaeger and Rossler, 2010), this might indicate
that enhancing insight using psycho-educational approaches and high
transparency when applying coercive practices could improve patients
appreciation of these procedures, and compensate for the negative ef-
fects of perceived coercion on treatment adherence.
One surprising finding was that perceived coercion and therapeutic
alliance were not associated in our model. It may be that while the
majority of patients have experienced trauma during their psychiatric
treatment, these experiences do not outweigh the benefits of the re-
lationship with their usual psychiatrist whereas a positive association
was found when exploring the therapeutic relationship with the
318
Psychiatry Research 257 (2017) 315–321
Fig. 1. Structural Equation Model (SEM). Structural Equation
Model (SEM) with insight (BIS), symptomatology (PANSS), per-
ceived trauma related to psychiatric treatment, therapeutic alli-
ance (4PAS) and medication adherence (MARS).
clinician most responsible of their admission (Sheehan and Burns,
2011). However, our finding is congruent with the results from Roche
et al. (Roche et al., 2014) where no differences were found on the
therapeutic relationship in the individuals who had experienced re-
straint, seclusion, or the use of medication without consent. Given the
lack of research on this topic to date no firm conclusion can be drawn.
4.1. Limits, perspectives and conclusions
First, a major methodological limitation is the small sample size.
Although, the final testing model showed a good fit based on four
statistical indices, it should be replicated in a larger sample. Second,
perceived trauma was retrospectively assessed and patients’ assessment
of the therapeutic relationship may have reflected an ongoing re-
lationship with the clinician during hospitalization. The questionnaire
was developed for this study. This assessment of subjective traumatic
experiences related to psychiatric care was not based on a validated
questionnaire. As this assessment reflected the endorsement of a single
item, it may have been subject to recall bias or influenced by the par-
ticipant's mental state. Third, insight was evaluated using the BIS scale
(Birchwood et al., 1994). This self-report insight scale (BIS) is a quick
and reliable measure exploring insight in a multidimensional concept
which include the subjective impression of patient facing to clinical
symptoms and two items on acceptance to drug treatments that may be
related to medication adherence (MARS). However, this chose could
constitute a limitation to explore the entire patient's subjective points of
view on the disease beyond the clinical condition with a potential im-
pact on therapeutic adherence. Future longitudinal studies may be
warranted to investigate this issue. Moreover, it would be useful to
establish whether the model reported here will be confirmed long-
itudinally.
Despite these limitations, the present study has several important
strengths. The current results confirm the important influence of ther-
apeutic alliance and insight on medication adherence. Moreover, to the
best of our knowledge, this is the first study exploring the mediating
role of perceived trauma on medication adherence using pathway
methodology. In terms of clinical implications, medication adherence
could be enhanced either by reducing perceived trauma or by in-
creasing insight. The present study articulates the need for mental
health clinicians to be sensitive to patients' perception of traumatic
experiences related to psychiatric treatment and to be aware of the
importance of therapeutic relationship in treating patients with schi-
zophrenia. International guidelines (Hasan et al., 2013) recommend to
incorporate patients’ perspectives and to encourage greater involve-
ment in care via shared decision-making results in greater adherence to
treatment, more effective disease self-management, and greater patient
 Table 3:
Observed pairwise correlation and covariance matrix of variables included in the study.*
A. Tessier et al.

319
Psychiatry Research 257 (2017) 315–321
A. Tessier et al.
satisfaction. Involving patients in a shared medical decision may reduce
the traumatic impact of psychiatric care and reinforce therapeutic en-
gagement in patients with schizophrenia. Finally, this study raises
questions that need further exploration to fully understand the complex
relationship between insight, therapeutic relationship, adherence and
outcomes.
Conflict of interest
None
Contributors
D. Misdrahi and F. Baylé designed the study. D. Misdrahi, F. Baylé
and P-M. Llorca carried out inclusion and collected data during an
extensive standardized evaluation. A. Tessier and L. Boyer performed
the statistical analysis. A. Tessier and D. Misdrahi wrote the first
complete manuscript. L. Boyer, M. Husky, F. Baylé and P-M. Llorca
provided substantial modification to the final manuscript. All authors
were involved in the project and approved the final manuscript.
Role of the funding source
This work was supported by a grant from the Clinical Research
Hospital Program from the French Ministry of Health (PHRC “2012”).
The funding source had no role in the conduct or publication of the
study.
Acknowledgments
None
References
Amador, X.F., Flaum, M., Andreasen, N.C., Strauss, D.H., Yale, S.A., Clark, S.C., Gorman,
J.M., 1994. Awareness of illness in schizophrenia and schizoaffective and mood
disorders. Arch. Gen. Psychiatry 51, 826–836.
American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental
Disorders, Fourth ed. American Psychiatric Pub (DSM-IV-TR®).
Ascher-Svanum, H., Faries, D.E., Zhu, B., Ernst, F.R., Swartz, M.S., Swanson, J.W., 2006.
Medication adherence and long-term functional outcomes in the treatment of schi-
zophrenia in usual care. J. Clin. Psychiatry 67, 453–460.
Baloush-Kleinman, V., Levine, S.Z., Roe, D., Shnitt, D., Weizman, A., Poyurovsky, M.,
2011. Adherence to antipsychotic drug treatment in early-episode schizophrenia: a
six-month naturalistic follow-up study. Schizophr. Res. 130, 176–181.
Beck, A.T., Steer, R.A., Carbin, M.G., 1988. Psychometric properties of the Beck
Depression Inventory: twenty-five years of evaluation. Clin. Psychol. Rev. 8, 77–100.
Beck, E.-M., Cavelti, M., Kvrgic, S., Kleim, B., Vauth, R., 2011. Are we addressing the
“right stuff” to enhance adherence in schizophrenia? Understanding the role of in-
sight and attitudes towards medication. Schizophr. Res. 132, 42–49.
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
Psychiatr. Scand. 89, 62–67.
Davies, G., Fowler, D., Greenwood, K., 2016. Metacognition as a mediating variable be-
tween neurocognition and functional outcome in first episode psychosis. Schizophr.
Bull.
Day, J.C., Bentall, R.P., Roberts, C., Randall, F., Rogers, A., Cattell, D., Healy, D., Rae, P.,
Power, C., 2005. Attitudes toward antipsychotic medication: the impact of clinical
variables and relationships with health professionals. Arch. Gen. Psychiatry 62,
717–724.
Drake, R.J., Nordentoft, M., Haddock, G., Arango, C., Fleischhacker, W.W., Glenthøj, B.,
Leboyer, M., Leucht, S., Leweke, M., McGuire, P., Meyer-Lindenberg, A., Rujescu, D.,
Sommer, I.E., Kahn, R.S., Lewis, S.W., 2015. Modeling determinants of medication
attitudes and poor adherence in early nonaffective psychosis: implications for in-
tervention. Schizophr. Bull. 41, 584–596.
Fond, G., Boyer, L., Boucekine, M., Aden, L.A., Schürhoff, F., Tessier, A., Andrianarisoa,
M., Berna, F., Brunel, L., Capdevielle, D., Chereau, I., Mallet, J., Denizot, H., Dorey,
J.M., Dubertret, C., Dubreucq, J., Faget, C., Gabayet, F., Rey, R., Richieri, R.,
Passerieux, C., Schandrin, A., Urbach, M., Vidailhet, P., Llorca, P.M., Misdrahi, D.,
FACE-SZ (FondaMental Academic Centers of Expertise for Schizophrenia) group,
2016. Validation study of the Medication Adherence Rating Scale. Results from the
FACE-SZ national dataset. Schizophr. Res.
Frank, A.F., Gunderson, J.G., 1990. The role of the therapeutic alliance in the treatment of
schizophrenia. Relationship to course and outcome. Arch. Gen. Psychiatry 47,
228–236.
320
Psychiatry Research 257 (2017) 315–321
Gilburt, H., Rose, D., Slade, M., 2008. The importance of relationships in mental health
care: a qualitative study of service users' experiences of psychiatric hospital admis-
sion in the UK. BMC Health Serv. Res. 8, 92.
Gilmer, T.P., Dolder, C.R., Lacro, J.P., Folsom, D.P., Lindamer, L., Garcia, P., Jeste, D.V.,
2004. Adherence to treatment with antipsychotic medication and health care costs
among Medicaid beneficiaries with schizophrenia. Am. J. Psychiatry 161, 692–699.
Gunzler, D., Chen, T., Wu, P., Zhang, H., 2013. Introduction to mediation analysis with
structural equation modeling. Shanghai Arch. Psychiatry 25, 390–394.
Hasan, A., Falkai, P., Wobrock, T., Lieberman, J., Glenthoj, B., Gattaz, W.F., Thibaut, F.,
Möller, H.-J., WFSBP Task force on Treatment Guidelines for Schizophrenia, 2013.
World Federation of Societies of biological psychiatry (WFSBP) guidelines for bio-
logical treatment of schizophrenia, part 2: update 2012 on the long-term treatment of
schizophrenia and management of antipsychotic-induced side effects. World J. Biol.
Psychiatry Off. J. World Fed. Soc. Biol. Psychiatry 14, 2–44.
Jaeger, M., Rossler, W., 2010. Enhancement of outpatient treatment adherence: patients'
perceptions of coercion, fairness and effectiveness. Psychiatry Res. 180, 48–53.
Kaltiala-Heino, R., Laippala, P., Salokangas, R.K., 1997. Impact of coercion on treatment
outcome. Int. J. Law Psychiatry 20, 311–322.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale
(PANSS) for schizophrenia. Schizophr. Bull. 13, 261–276.
Lacro, J.P., Dunn, L.B., Dolder, C.R., Leckband, S.G., Jeste, D.V., 2002. Prevalence of and
risk factors for medication nonadherence in patients with schizophrenia: a compre-
hensive review of recent literature. J. Clin. Psychiatry 63, 892–909.
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. Schizophr. Res. 102, 295–302.
Lecrubier, Y., Sheehan, D.V., Weiller, E., Amorim, P., Bonora, I., Sheehan, K.H., Janavs,
J., Dunbar, G.C., 1997. The Mini International Neuropsychiatric Interview (MINI). A
short diagnostic structured interview: reliability and validity according to the CIDI.
Eur. Psychiatry 12, 224–231.
Lincoln, T.M., Lüllmann, E., Rief, W., 2007. Correlates and long-term consequences of
poor insight in patients with schizophrenia. A systematic review. Schizophr. Bull. 33,
1324–1342.
Lingjaerde, O., Ahlfors, U.G., Bech, P., Dencker, S.J., Elgen, K., 1987. The UKU side effect
rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-
sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr. Scand.
Suppl. 334, 1–100.
Llorca, P.-M., 2008. Partial compliance in schizophrenia and the impact on patient out-
comes. Psychiatry Res. 161, 235–247.
McCabe, R., Bullenkamp, J., Hansson, L., Lauber, C., Martinez-Leal, R., Rössler, W.,
Salize, H.J., Svensson, B., Torres-Gonzalez, F., van den Brink, R., Wiersma, D., Priebe,
S., 2012. The therapeutic relationship and adherence to antipsychotic medication in
schizophrenia. PLoS One 7, e36080.
Mielau, J., Altunbay, J., Gallinat, J., Heinz, A., Bermpohl, F., Lehmann, A., Montag, C.,
2015. Subjective experience of coercion in psychiatric care: a study comparing the
attitudes of patients and healthy volunteers towards coercive methods and their
justification. Eur. Arch. Psychiatry Clin. Neurosci.
Misdrahi, D., Petit, M., Blanc, O., Bayle, F., Llorca, P.-M., 2012. The influence of ther-
apeutic alliance and insight on medication adherence in schizophrenia. Nord. J.
Psychiatry 66, 49–54.
Misdrahi, D., Tessier, A., Swendsen, J., Berna, F., Brunel, L., Capdevielle, D., Chereau, I.,
Danion, J.-M., De Pradier, M., Dorey, J.-M., Dubertret, C., Dubreucq, J., Faget, C.,
Gabayet, F., Rey, R., Richieri, R., Passerieux, C., Schandrin, A., Schurhoff, F.,
Tronche, A.M., Urbach, M., Llorca, P.M., Fond, G., FACE-SZ (FondaMental Academic
Centers of Expertise for Schizophrenia) Group, 2016. Determination of adherence
profiles in schizophrenia using self-reported adherence: results from the FACE-SZ
dataset. J. Clin. Psychiatry 77, e1130–e1136.
Misdrahi, D., Verdoux, H., Lançon, C., Bayle, F., 2009. The 4-Point ordinal Alliance Self-
report: a self-report questionnaire for assessing therapeutic relationships in routine
mental health. Compr. Psychiatry 50, 181–185.
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. Schizophr. Bull. 35,
336–346.
Muthén, L.K., Muthén, B.O., 1998. Mplus User’s Guide. Sixth Edition. Los Angel. CA.
Novick, D., Montgomery, W., Treuer, T., Aguado, J., Kraemer, S., Haro, J.M., 2015.
Relationship of insight with medication adherence and the impact on outcomes in
patients with schizophrenia and bipolar disorder: results from a 1-year European
outpatient observational study. BMC Psychiatry 15, 189.
Paksarian, D., Mojtabai, R., Kotov, R., Cullen, B., Nugent, K.L., Bromet, E.J., 2014.
Perceived trauma during hospitalization and treatment participation among in-
dividuals with psychotic disorders. Psychiatr. Serv. Wash. DC 65, 266–269.
Robinson, D., Woerner, M.G., Alvir, J.M., Bilder, R., Goldman, R., Geisler, S., Koreen, A.,
Sheitman, B., Chakos, M., Mayerhoff, D., Lieberman, J.A., 1999. Predictors of relapse
following response from a first episode of schizophrenia or schizoaffective disorder.
Arch. Gen. Psychiatry 56, 241–247.
Rocca, P., Crivelli, B., Marino, F., Mongini, T., Portaleone, F., Bogetto, F., 2008.
Correlations of attitudes toward antipsychotic drugs with insight and objective psy-
chopathology in schizophrenia. Compr. Psychiatry 49, 170–176.
Roche, E., Madigan, K., Lyne, J.P., Feeney, L., O’Donoghue, B., 2014. The therapeutic
relationship after psychiatric admission. J. Nerv. Ment. Dis. 202, 186–192.
Samalin, L., de Chazeron, I., Blanc, O., Brunel, L., Fond, G., Llorca, P.-M., 2016. Attitudes
toward antipsychotic medications as a useful feature in exploring medication non-
adherence in schizophrenia. Schizophr. Res. 178, 1–5.
Sheehan, K.A., Burns, T., 2011. Perceived coercion and the therapeutic relationship: a
neglected association? Psychiatr. Serv. Wash. DC 62, 471–476.
A. Tessier et al.
Thomas, M.L., Green, M.F., Hellemann, G., Sugar, C.A., Tarasenko, M., Calkins, M.E.,
Greenwood, T.A., Gur, R.E., Gur, R.C., Lazzeroni, L.C., Nuechterlein, K.H., Radant,
A.D., Seidman, L.J., Shiluk, A.L., Siever, L.J., Silverman, J.M., Sprock, J., Stone, W.S.,
Swerdlow, N.R., Tsuang, D.W., Tsuang, M.T., Turetsky, B.I., Braff, D.L., Light, G.A.,
2017. Modeling deficits from early auditory information processing to psychosocial
functioning in schizophrenia. JAMA Psychiatry 74, 37–46.
Velligan, D.I., Weiden, P.J., Sajatovic, M., Scott, J., Carpenter, D., Ross, R., Docherty, J.P.,
Expert Consensus Panel on Adherence Problems in Serious and Persistent Mental
Illness, 2009. The expert consensus guideline series: adherence problems in patients
321
Psychiatry Research 257 (2017) 315–321
with serious and persistent mental illness. J. Clin. Psychiatry 70 (Suppl 4), 1–46 (quiz
47-48).
Weiden, P.J., Kozma, C., Grogg, A., Locklear, J., 2004. Partial compliance and risk of
rehospitalization among California Medicaid patients with schizophrenia. Psychiatr.
Serv. Wash. DC 55, 886–891.
Wittorf, A., Jakobi, U., Bechdolf, A., Müller, B., Sartory, G., Wagner, M., Wiedemann, G.,
Wölwer, 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. Eur. Psychiatry J. Assoc. Eur. Psychiatr. 24, 259–267.