Journal of Behavior Therapy and Experimental Psychiatry
A comparative study of patients and therapists' reports of schema modes
Introduction
Next to persistent antisocial behaviour, one of the main features of antisocial personality disorder (ASPD) is deceitfulness as manifested in repeated lying (American Psychiatric Association, 2005). Additionally, antisocial patients are characterized by defensive responding (de Ruiter & Greeven, 2000) and a tendency to over-report healthy behaviour (Cima, 2003). This response style of antisocial patients forms a major problem because it diminishes the reliability of their self-report, which has negative effects for both therapy and the reliability of research. It has also caused several authors to advise against the use of self-report inventories with forensic subjects (Gacono and Meloy, 1994, Hare, 1991) and probably contributed to this patient group being viewed as therapy-resistant (Harris & Rice, 2006).
One possible way to circumvent the denying response style of antisocial patients is by using alternative sources of information than self-report. Since the therapeutic relationship is considered a useful context for assessing key dysfunctional beliefs (Beck et al., 2001), this study compares self-report by patients with ASPD, borderline personality disorder (BPD) and cluster C personality disorder (ClC-PD) with reports by these patients' therapists. This way, it can be assessed whether the discrepancy between patients and therapists is specific for patients with an ASPD as compared to other PD patients.
We asked patients and therapists to rate schema modes, one of the central concepts of Schema-Focused Therapy (SFT, Young, Klosko, & Weishaar, 2003). Schema modes reflect state-depending clusters of thoughts, feelings and behaviours. Recently, it was demonstrated that SFT was highly effective in treating borderline patients (Giesen-Bloo et al., 2006). Furthermore, SFT is becoming increasingly implemented within forensic treatment settings (Bernstein, Arntz, & de Vos, 2007). Modes can be adaptive or maladaptive. Until now, 14 different schema modes have been identified that can be clustered into four categories; first, maladaptive child modes that result out of unmet core childhood needs, second, dysfunctional coping modes that correspondent to an overuse of the fight, flight or freeze coping styles and third, dysfunctional parent modes that reflect behaviour of the patients' parent(s) towards the patient as a child that the patient has internalized. Fourth, there are two healthy modes; that of the Healthy and that of the Happy Child (for an overview of the modes, see Lobbestael, van Vreeswijk, & Arntz, 2008).
A previous study (Lobbestael, Arntz, & Sieswerda, 2005) already raised questions about the reliability of self-reported schema modes in antisocial patients because these patients indicated very high levels of healthy modes that even did not differ significantly from non-patient controls, and surprisingly low maladaptive modes. In another study (Lobbestael et al., 2008), ASPD was negatively correlated to several maladaptive self-reported modes. Cleary, these findings do not match clinical observation of high levels of pathology in ASPD.
We are not aware of any previous studies comparing self- and other-report of schema modes or other cognitive concepts in specific PDs. In sum, the present study compares the self-reported schema modes of PD patients with the mode ratings of their therapists. We hypothesize that there will be a strong discrepancy between self- and other report in ASPD with relatively underreporting of maladaptive constructs by these patients. More specifically, we expect therapists to indicate a higher level of pathological modes and a lower level of adaptive modes than the antisocial patients report themselves. In contrast, more agreement is expected between self-and other report of modes in patients with BPD and ClC-PD, which are used as PD control groups. Additionally, the influence of the level of psychopathy of the antisocial group on the self-versus other report will be tested. In this way the hypothesis will be tested that antisocial patients that are high in psychopathy would be even more prone to relatively underreport maladaptive modes than ASPD-patients low in psychopathy.
Section snippets
Subjects
Self-reported modes were compared with mode report by their therapists for N = 92 patients, divided over three patient groups: patients with ASPD (n = 18), patients with BPD (n = 47) and patients with ClC-PD (avoidant, dependent and/or obsessive-compulsive PD, n = 27). Four participants had to be removed because of missing data. The antisocial patients were all male, while the borderline group consisted of 38 women and 11 men, and the ClC-PD group of 20 women and 8 men. Consequently, the groups
Patient versus therapist ratings
Means of the self- and other mode scores for each group are presented in Table 1. Repeated measures analyses revealed a significant three-way interaction for self-other report × modes × group for the maladaptive modes, F (22, 86) = 2.49, p = .001, but not for the adaptive modes, F (2,89) = .20, p = .82. Results of the t-tests assessing whether self- and other-report differed significantly, are presented in Table 2 for each group. In the ASPD group, self- and other-reported modes differed significantly in
Discussion
In line with our expectation, patients with ASPD rated the presence of most of their maladaptive modes markedly lower compared to their therapists. This discrepant pattern was only observed for some of the modes in the borderline and ClC-PD group. Furthermore, the patients-therapists discrepancy was significantly stronger in the antisocial group than in the two PD control groups. This strong discrepancy in maladaptive mode rating of the ASPD-patients and their therapists can be interpreted at
Acknowledgements
Thanks are due to Anke Verstegen, Christine van Giesen, Minda Dijkstra, Nelly Dorssers, Tamara Schrijvers and Yvette Haenen for their help in collecting the data. We are grateful for the collaboration of the direction board, staff and of patients the ‘Rooyse Wissel’ in Venray; the ‘RIAGG Maastricht’; GGZ Midden Brabant and Midden Limburg; ‘Mutsaersoord’ in Venray; the Symfora group, location Amerfoort; the ‘Pompekliniek’ in Nijmegen; ‘Lionarons’ in Heerlen; and the ‘Viersprong’ in Halsteren;
References (32)
- et al.
An experimental test of the schema mode model of borderline personality disorder
Journal of Behavior Therapy and Experimental Psychiatry
(2005) - et al.
Dysfunctional beliefs discriminate personality disorders
Behaviour Research and Therapy
(2001) - et al.
Schema modes and childhood abuse in borderline and antisocial personality disorders
Journal of Behavior Therapy and Experimental Psychiatry
(2005) - et al.
An empirical test of schema mode conceptualizations in personality disorders
Behaviour Research and Therapy
(2008) - et al.
Inter-rater reliability of the SCID alcohol and substance use disorder section among adolescents
Drug and Alcohol Dependence
(2000) - et al.
Attainment and maintenance of reliability of axis I and II disorders over the course of a longitudinal study
Comprehensive Psychiatry
(2001) Diagnostic and statistical manual of mental disorders
(2005)- et al.
Schema focused therapy in forensic settings: theoretical model and recommendations for best clinical practice
International Journal of Forensic Mental Health
(2007) - et al.
A multigroup item response theory analysis of the psychopathy checklist-revised
Psychological Assessment
(2004) Faking good, bad and ugly. Malingering in forensic psychiatry inpatients
(2003)