Le questionnaire abrégé des schémas (YSQ-S2) proposé par Young (1998) [13Young J.E. Young Schema Questionnaire Short Form  New York, NJ: Cognitive Therapy Center (1998). 

Cliquez ici pour aller à la section Références] est un instrument bref (75 items) permettant de mesurer 15 schémas précoces inadaptés (SPI). L’objectif de cette étude est de valider empiriquement deux versions du YSQ-S2 (version française initiale et version française modifiée) auprès d’un échantillon de patients présentant un trouble de la personnalité borderline (n=37) et de sujets témoins (n=263). Les résultats indiquent que les deux versions de l’instrument distinguent bien les patients borderline des sujets témoins. Les comparaisons statistiques montrent des différences significatives entre les deux groupes, les scores au YSQ-S2 étant à chaque fois plus élevés pour le groupe pathologique. Pour les deux versions, un seuil pathologique est également proposé. La version française (initiale ou modifiée) du YSQ-S2 apparaît être une mesure valide permettant la mise en évidence des SPI et une bonne échelle de mesure du trouble de la personnalité borderline.

According to Young [11Young J.E. Cognitive therapy for personality disorders: a schema focused approach  Sarasota, FL: Professional Resource Press (1990). 

Cliquez ici pour aller à la section Références], Early Maladaptive Schemas (EMS) are deep cognitive structures constituted by beliefs about the word, oneself, and one’s relationship with others. Young proposed a first clinical scale to assess EMS: the Young Schema Questionnaire (YSQ-L1). A later version, the YSQ-L2 (205 items), was developed by Young to measure 16 EMS. The Schema Questionnaire-Short Form (YSQ-S2) was designed to measure 15 EMS [13Young J.E. Young Schema Questionnaire Short Form  New York, NJ: Cognitive Therapy Center (1998). 

Cliquez ici pour aller à la section Références] and is a shorter instrument (75 items).

The aim of the present study was to validate the French version of the YSQ-S2 through a comparison of patients with borderline personality disorder (BPD) with control subjects. We used two French versions of the YSQ-S2 for validation purposes in BPD and control samples. The first version (initial French version) is identical to the original YSQ-S2 (75 items, 15 EMS) while the second (modified French version) comprises 68 items grouped into 14 factors.

The control group was composed of 263 non-clinical subjects (82 males, 181 females) who were mostly university students. The mean age of the sample was 27.92 years (SD=14.26) and age ranged from 17 to 67 years. The pathological group was composed of 37 BPD patients (eight males and 29 females) from the Anxiety Disorder Unit (Neurological Hospital, Lyon) and the Cognitive Behaviour Therapy Unit (Sainte Marguerite Hospital, Marseille). The majority of these patients were employees and senior executives. Age ranged from 19 to 53 years, with an average of 34.45 years (SD=9.74). A psychiatrist diagnosed the patients with the Diagnostic Interview for Borderline-Revised [3Chaine F., Guelfi J.D., Monier C., et al. Diagnostic clinique et évaluation standardisée de la personnalité borderline : rapport préliminaire Encéphale 1995 ;  21 : 247-256

Cliquez ici pour aller à la section Références, 17Zanarini M.C., Gunderson J.G., Frankenburg F.R., et al. The revised diagnostic interview for borderlines: discriminating BPD from other axis II disorders J Personal Disord 1989 ;  3 : 10-18 [cross-ref]

Cliquez ici pour aller à la section Références]. All participants filled in the French version of the original 205-item YSQ-L2 [5Cottraux J., Blackburn I.M. Psychothérapies cognitives des troubles de la personnalité  Paris: Masson (2006). 

Cliquez ici pour aller à la section Références] from which responses of the 75-item YSQ-S2 were extracted. Control subjects anonymously completed the YSQ-L2 in groups of 10 to 40 people and patients completed the YSQ-L2 in a clinical setting.

For each version of the YSQ-S2, the total score was analyzed with two-ways ANOVA (Group×Gender) and the sub-scores were analyzed with one-way MANOVA (Group). Our results showed that the two versions of the YSQ-S2 have good discrimination values between BPD patients and control subjects. Statistical comparisons indicated significant differences between the two groups. No difference appeared between males and female’s scores. Results showed that BPD patients’ total score and sub-scores were significantly higher than those of control subjects. Moreover, for each version of the YSQ-S2, a threshold level of pathology was obtained by the analysis of the intervals of total scores for the two groups. In the initial French version of the YSQ-S2, the scores of 15.59% of the control subjects overlapped with the scores of 94.59% of the BPD patients with a total score higher than 180. In the modified French version, the scores of 15.97% of the control subjects overlapped with the scores of 94.59% of the BPD patients with a total score higher than 160. Therefore, the pathological threshold of BPD could be set at a score of 181 and 161, respectively for the initial French version and the modified French version of the YSQ-S2.

In spite of the small size of our pathological sample (N=37), the French version (initial or modified) of the YSQ-S2 appears to be a valid measurement allowing the description of the EMS and a quite effective instrument for measuring BPD symptoms. In addition, our results suggest that it is possible to reduce the YSQ-S2, given that the suppression of several items (modified French version of the questionnaire) does not modify the cut-off point and the differences between the BPD patients and the control subjects.