Acceptance and Commitment Therapy for Management of Suicidal Patients: A Pilot Study

Abstract

(6) pharmacological treatment and number of visits for psychiatric emergencies in the previous 3 months. Self-assessments included: (1) suicidal ideation on a visual analogue scale from 0 (none) to 10 (maximum), currently and during the last 15 days; (2) depressive intensity using Beck Depression Inventory-II (BDI); (3) psycho-logical pain on a visual analogue scale from 0 (none) to 10 (maxi-mum); (4) anxiety state using the State-Trait Anxiety Inventory (STAI); (5) hopelessness during the last month using the Beck Hopelessness Scale; (6) quality of life using the World Health Or-ganization Quality of Life measure, and (7) acceptance using the Acceptance and Action Questionnaire (AAQ) [10] . Considering the small sample size and skewed distribution of most continuous variables, median (min–max) and non-paramet-ric tests were used. Friedman’s tests for quantitative variables and Cochran’s Q test for categorical variables were used to globally compare the three time points. If the global p value was p < 0.05, two-by-two comparisons were run (signed rank, McNemar’s and kappa tests) and corrected for multiple comparisons (Bonferroni correction). Spearman’s rank order correlations were applied to measure associations between continuous variables. Statistical analyses were performed using SAS software, version 9.2 (SAS In-stitute, Cary, N.C., USA). Two patients did not start the programme; 78.8% attended all sessions. ACT was reported to be helpful by 96.9% patients. One patient (suffering from borderline personality disorder) commit-ted suicide during the first month of the study, without any clear association with intervention. Sociodemographic characteristics were as follows: 57.1% males, median age 38.4 years (min–max: 18–60), 48.6% single, 60% fully employed. The majority of patients were currently depressed (94.3%), having attempted suicide the month preceding the inclu-sion (93.9%). Psychiatric disorders were unipolar (37.1%), bipolar (57.1%), anxiety (97.1%), addictive (28.6%), eating (17.1%), and/or borderline personality (20%) disorders. All participants were on psychotropic medications. There were significant differences for all scores between the three visits (p < 0.001). Between inclusion and the 1-week follow-up, there was a significant reduction of the C-SSRS ‘suicidal ide-ation’ subscore [20 (0–30) vs. 0 (0–20), respectively; p < 0.001], SSI score [7 (0–22) vs. 0 (0–10); p < 0.001] as well as intensity of current and previous suicidal ideations during the last 15 days [1 (0–10) vs. 0 (0–3), 2 (0–9) vs. 0 (0–5), respectively; both p < 0.001]. Of note, reduction in the C-SSRS ‘suicidal ideation’ subscore was cor-related to the AAQ score (p = 0.04, r = –0.37), but not to BDI and STAI scores. Between inclusion and 3-month follow-up, the re-duction of all suicidal ideation scores remained significant. There were no suicide reattempts during the follow-up period. With one million deaths every year, suicide is a major health problem worldwide. Recently, suicidal behaviour disorder (SBD) has been included in DSM-5 as an independent clinical entity [1] . It highlights the need to address the suicidal process as a primary target of concern and to identify corresponding transnosographic preventive strategies, including psychotherapy. Acceptance and Commitment Therapy (ACT), a ‘third-wave’ behavioural therapy, targets experiential avoidance (tendency to avoid unwanted thoughts or emotions) at the core of psychiatric disorders. Of note, suicidal subjects report intrusive mental images of suicide that they try to suppress, increasing in their intensity and frequency, inde-pendently from depressive symptoms [2] . Indeed, experiential avoidance would predict suicidal behaviours [3] . Moreover, two case reports have suggested the preventive role of ACT on suicidal reattempts at 1 year [4] . Thus, we aimed at examining the useful-ness of an add-on ACT group programme to decrease suicidal ide-ation in high-risk patients. The study was conducted in the Department of Psychiatric Emergency and Acute Crisis, Academic Hospital of Montpellier, France. Thirty-five outpatients suffering from a current SBD ac-cording to DSM-5 [1] (history of suicidal attempt in the past year) were included in an ACT programme ( table 1 ) [5–7] . Exclusion criteria were current mania or depressive episode with mixed fea-tures, and schizophrenia. Patients were assessed at inclusion (1 week before the pro-gramme, T0) and 1 week (T1) and 3 months (T2) after programme completion. Psychiatrists assessed: (1) suicidal ideation using the Columbia-Suicide Severity Rating Scale (C-SSRS) [8] (suicidal ide- ation subscore = severity and intensity items) and the Scale for Suicidal Ideation (SSI score) [9]; (2) psychiatric disorders using the French version of the Mini International Neuropsychiatric Inter-view; (3) borderline personality disorder using the Screening In-terview for Axis II Disorder; (4) depressive symptomatology using the Inventory of Depressive Symptomatology (IDS-C30); (5) glob-al functioning using the Functioning Assessment Short Test, and