Effets de la pratique de la pleine conscience sur la santé mentale des enseignants : une étude pilote contrôlée non randomisée

Abstract

Cette étude pilote examine la faisabilité et les bénéfices d’un programme de 18 h de pleine conscience (mindfulness) MBSR/MBCT proposé à 79 enseignants français d’écoles primaires, de collèges et de lycées, dont 52 en groupe expérimental, 27 en groupe témoin, organisé sur six ou sept semaines à l’École supérieure du professorat et de l’éducation (ESPE) de l’université de Limoges. Un protocole quasi expérimental, contrôlé mais non randomisé, avec groupe témoin en liste d’attente. Les bénéfices ont été évalués trois fois (pré-/post-traitement, suivi à deux mois) sur huit échelles explorant les difficultés de santé mentale et la santé mentale positive : questionnaire général de santé (GHQ-28), échelle de stress perçu (PSS-14), double échelle HAD anxiété-dépression, échelle d’auto-efficacité, échelle de satisfaction de vie (SWLS), échelle de bien-être (WBQ-12), questionnaire de mindfulness (FMI-14). Les résultats montrent des bénéfices significatifs pré-/post-traitement sur les huit échelles, qui se maintiennent dans le suivi à deux mois. La recherche d’éventuelles corrélations entre le temps de pratique personnelle de pleine conscience effectué à domicile et les bénéfices recueillis n’a donné de résultats que sur l’échelle FMI-14. Enfin, des corrélations sont observées entre l’augmentation pré-/post-intervention des progrès en mindfulness (FMI-14) et les bénéfices acquis sur les sept autres échelles.

Introduction

This pilot study explores the feasibility and benefits of a mindfulness MBSR/MBCT (Mindfulness-Based Stress Reduction, Mindfulness-Based Cognitive Therapy) program, which was offered to 79 French teachers from the Académie de Limoges who were enrolled in Ministry of National Education's training courses. The teaching profession is highly exposed to stress and related mental health problems, such as anxiety, depression and even burnout. The pressure of classroom organization, the continuous tension of concentration, the management of student behavior, the relationships with parents, the lack of support from the administration can all greatly affect the physical and mental health of teachers. Mindfulness-Based Interventions (MBIs) developed from the MBSR and MBCT programs have shown significant benefits for teachers in reducing stress, reducing associated psychological symptoms and improving well-being.

Method

Seventy-nine volunteer teachers from primary, middle and high schools participated in all six or seven sessions of an 18 hours MBSR/MBCT internship (3 h or 2 h 35 per week), fifty-two in an experimental group, twenty-seven in the control group, from 2012 to 2015 (n = 67 women, 85%). A quasi-experimental, controlled but non-randomized protocol was used, with a control group on a waiting list. Benefits were assessed three times (pre-/post-treatment, two-month follow-up) using eight scales, four assessing mental health difficulties (general health questionnaire GHQ-28, perceived stress scale PSS-14, hospital anxiety and depression scale HAD), four scales assessing positive mental health (self-efficacy scale, satisfaction with life scale SWLS, well-being scale WBQ-12, dispositional mindfulness FMI-14). Ninety-seven questions were asked. The eight scales were administered at the same time to all experimental and control groups (early January in pre-test, late February in post-test), then again two months later (late April) by the experimental groups. The internships were conducted by the first author of this article. The seventy-nine teachers selected in the research participated in all the sessions (or remedial sessions offered during the week to absentees) and returned all the questionnaires anonymously.

Results and discussion

Variance analyses show that the mindfulness and control groups are homogeneous in the pre-test on the eight scales. Comparisons performed with the Student's t-test reveal significant pre-/post-interventional improvements (P < 0.01) in the experimental mindfulness group across all scales of mental health difficulties and positive mental health. For mental health difficulties, the effect sizes (Cohen's d) are high (d ≥ 0.80) for general health GHQ-28 and perceived stress PSS-14, medium (d ≥ 0.50) for anxious and depressive HAD symptoms. For positive mental health, effect sizes are quite large (d = 0.74) for well-being WBQ-12, medium (d ≥ 0.50) for dispositional mindfulness FMI-14, low (d ≥ 0.20) for self-efficacy and life satisfaction SWLS. All these improvements in mental health difficulties and positive mental health are generally maintained in the two-month follow-up. The control group maintains stable scores (P > 0.5) on the different scales except for anxiety HAD and depression HAD, where a slight reduction in anxious symptomatology (d = 0.28) and depressive symptomatology (d = 0.41) are observed, possibly due to a calendar effect as mood may become lighter as the February holidays were getting closer during the post-test. We note a fairly high observance of the “formal practice” time, which was carried out at home daily by each teacher, five days a week during the internship, for an average of 21.3 min. However, the search for possible correlations between the practice time and the benefits collected on the eight scales only yielded significant results (P < 0.01) on the FMI-14 scale, with a moderate correlation (r = 0.35). The study of such correlations gives very heterogeneous results in MBIs research (Shapiro et al., 2007; Ruths et al., 2013; Crane et al., 2014; Carmody and Baer, 2008; Flook et al., 2013; Vettese et al., 2009; Parsons et al., 2017). Consequently, it has been considered that the effects might only appear from a critical threshold of daily practice time; thus, Carson et al. (2004) observe correlations starting from 32 min per day. Perhaps the 21.3 minutes recorded during our internship did not reach this critical threshold? It should be noted that the 52 teachers in the experimental group gave an importance score of 7.08/10 to the mindfulness internship. Finally, significant correlations (P < 0.0001), medium or high, were observed between the pre-/post-intervention increase in mindfulness progress (FMI-14) and the benefits acquired on the other scales (except on the SWLS scale, where the correlation is low at P < 0.016); this is consistent with the results of publications that highlight the central role of increased mindfulness disposition in the benefits provided by MBIs programs (Brown et al., 2003; Brown et al., 2007; Carmody and Baer, 2008; Carmody and Baer, 2009; Flook et al., 2013; Rupprecht et al., 2017).