Effectiveness of Acceptance and Commitment Therapy in treating depression and suicidal ideation in Veterans

Behaviour Research and Therapy 74 (2015) 25e31

Contents lists available at ScienceDirect

Behaviour Research and Therapy journal homepage: www.elsevier.com/locate/brat

Effectiveness of Acceptance and Commitment Therapy in treating depression and suicidal ideation in Veterans* Robyn D. Walser a, b, *, Donn W. Garvert a, Bradley E. Karlin c, d, Mickey Trockel e, f, Danielle M. Ryu e, f, C. Barr Taylor e, f a

National Center for PTSD, Veterans Affairs Palo Alto Health Care System, USA Department of Psychology, University of California, Berkeley, USA Mental Health Services, U.S. Department of Veterans Affairs Central Office, USA d Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University, USA e Department of Psychiatry, Stanford University Medical Center, USA f VISN 21 Mental Illness Research, Education and Clinical Center, Veterans Affairs Palo Alto Health Care System, USA b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 11 September 2014 Received in revised form 26 August 2015 Accepted 27 August 2015 Available online 31 August 2015

Objective: This paper examines the effects of Acceptance and Commitment Therapy for depression (ACTD), and the specific effects of experiential acceptance and mindfulness, in reducing suicidal ideation (SI) and depression among Veterans. Method: Patients included 981 Veterans, 76% male, mean age 50.5 years. Depression severity and SI were assessed using the BDI-II. Experiential acceptance and mindfulness were measured with the Acceptance and Action Questionnaire-II (AAQ-II) and the Five Facet Mindfulness Questionnaire, respectively. Results: Of the 981 patients, 647 (66.0%) completed 10 or more sessions or finished early due to symptom relief. For Veterans with SI at baseline, mean BDI-II score decreased from 33.5 to 22.9. For Veterans with no SI at baseline, mean BDI-II score decreased from 26.3 to 15.9. Mixed models with repeated measurement indicated a significant reduction in depression severity from baseline to final assessment (b ¼ 10.52, p < .001). After adjusting for experiential acceptance and mindfulness, patients with SI at baseline demonstrated significantly greater improvement in depression severity during ACT-D treatment, relative to patients with no SI at baseline (b ¼ 2.81, p ¼ .001). Furthermore, increases in experiential acceptance and mindfulness scores across time were associated with a reduction in depression severity across time (b ¼ 0.44, p < .001 and b ¼ 0.09, p < .001, respectfully), and the attenuating effect of mindfulness on depression severity increased across time (b ¼ 0.05, p ¼ .042). Increases in experiential acceptance scores across time were associated with lower odds of SI across time (odds ratio ¼ 0.97, 95% CI [0.95, 0.99], p ¼ .016) and the attenuating effect of experiential acceptance on SI increased across time (odds ratio ¼ 0.96, 95% CI [0.92, 0.99], p ¼ .023). Overall the number of patients with no SI increased from 44.5% at baseline to 65% at follow-up. Conclusions: Veterans receiving ACT-D demonstrated decreased depression severity and decreased odds of SI during treatment. Increases in experiential acceptance and mindfulness scores were associated with reduction in depression severity across time and increases in experiential acceptance scores were associated with reductions in SI across time. © 2015 Published by Elsevier Ltd.

Keywords: Acceptance and Commitment Therapy Suicide ideation Depression Veterans

Suicide is a complex behavior with far-reaching impact. Providing support and treatment to those suffering from suicidal

* Authors' note: This project was supported by Mental Health Services, U.S. Department of Veterans Affairs Central Office. * Corresponding author. National Center for PTSD, 334 e NCPTSD, VA Palo Alto Health Care System, 795 Willow Road, Menlo Park, CA 94804, USA. E-mail address: [email protected] (R.D. Walser).

http://dx.doi.org/10.1016/j.brat.2015.08.012 0005-7967/© 2015 Published by Elsevier Ltd.

ideation and behavior is a key mental health care priority. Veterans account for an estimated 20% of suicide deaths in the United States (Chakravorty et al., 2013), and the suicide rate among younger Veterans is rising (Institute of Medicine, 2010; Kuehn, 2009). As part of its important focus on suicide prevention, the Veterans Health Administration (VHA), the health care component of the U.S. Department of Veterans Affairs (VA), has been working to ensure veteran access to high quality mental health services and has

26

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31

implemented a number of specific suicide prevention strategies (Katz, 2012). Although a number of factors influence suicide ideation and attempts, depression is one of the greatest risk factors (Brown, Beck, Steer, & Grisham, 2000; Gotlib & Hammen, 2002). Reductions in depression have been associated with reductions in suicide ideation (Bruce et al., 2004; Mann et al., 2005). To improve the treatment of depression and other mental and behavioral health conditions, VHA has implemented a number of national evidence-based psychotherapy dissemination initiatives, which include competency-based training programs for each of these therapies (Karlin & Cross, 2014). One of these initiatives focuses on Acceptance and Commitment Therapy (ACT; Hayes, Strosahl, & Wilson, 2012) for depression (ACT-D; Walser, Sears, Chartier, & Karlin, in press; Zettle, 2007). Initial program evaluation data have shown that ACT-D resulted in significant overall reductions in depression (d ¼ 1.04) and improvements in quality of life among veterans treated by VA mental health providers being trained in ACT-D (Walser, Karlin, Trockel, Mazina, & Taylor, 2013). ACT utilizes a number of process interventions that may reduce suicidal ideation, as well as depression. One psychological process suggested as a pathway to suicidal ideation and suicide attempts is experiential avoidance (Chiles & Strosahl, 2005; Luoma & Villatte, 2012; Zettle, 2007) e the desire and actions taken to eliminate painful or unwanted thoughts and emotions. From this perspective, thinking about, planning or attempting suicide can be viewed as a way to solve the problem of intractable difficult emotional experience for the suicidal individual, and may be a strategy that is more likely to be used by those who have low tolerance for distress or ability to cope (Chiles & Strosahl, 2005). In a meta-analytic review (Hayes, Luoma, Bond, Masuda, & Lillis, 2006), experiential avoidance was purported to account for 16e25% of the variance in behavioral health problems including suicidal behavior. Additionally, in a recent evaluation, veterans demonstrated significant improvement in experiential acceptance (i.e., decrease in experiential avoidance) and mindfulness (see Walser et al., 2013). ACT expressly targets experiential avoidance using several core processes designed to decrease ineffective and problematic escape from emotional pain (e.g., suicide ideation and suicide attempts) by increasing experiential acceptance and mindful awareness. Four of the core processes used in ACT (i.e., present moment awareness, defusion, willingness and self-as-context; see Hayes et al., 2012) foster these outcomes. A number of studies have shown that promoting mindful awareness of psychological events decreases experiential avoidance (see Hayes et al., 2006 for a review) and increases psychological flexibility, a strong correlate of mental health (Kashdan & Rottenberg, 2010). Although data from clinical research studies, summarized in € rde Sive Vo € rding, and Hayes et al. (2006), Powers, Zum Vo Emmelkamp (2009), and Ruiz (2010), and real-world effectiveness evaluations have demonstrated the efficacy and effectiveness of ACT, there has been limited research on the effect of ACT on suicidal ideation and suicidal behavior (Luoma & Villatte, 2012). Moreover, the relationship between increases in experiential acceptance and mindful awareness, two key goals of ACT-D, and depression and suicidal ideation has not been examined in a veteran population. The purpose of this evaluation is to examine the effectiveness of ACT-D in reducing depression severity and suicidal ideation among veterans treated in routine clinical settings and to compare change in depression severity achieved by veterans with suicidal ideation at baseline with change in depression severity achieved by those without suicidal ideation at baseline, based on program evaluation data now available through the VA ACT-D Training Program. This evaluation also examines the association between depression

severity and suicide ideation during ACT-D treatment. In addition, the association between key ACT-D intervention process targets (experiential acceptance and mindfulness) and outcome variables (depression and suicidal ideation) are examined. 1. Methods 1.1. Program description The ACT for Depression treatment protocol (Walser et al., in press) was developed specifically for veterans and is intended to be administered in approximately 12e16 individual psychotherapy sessions. The protocol provides ACT-D specific information including (1) behavioral theory and background (e.g., the role of language in human suffering), (2) implementation of the six core processes (e.g., defusion, acceptance, present moment, self-ascontext, committed action and values; includes metaphors, exercises, example patient/therapist dialog), (3) specific patient skills (e.g., mindfulness, goals versus values distinction), (4) therapeutic alliance building (e.g., mindfulness rationale, use of compassion), (5) patient homework assignments and (6) useful appendices (e.g., safety planning worksheets, internet and hotline information). The protocol and training methods have been described in detail elsewhere (Walser et al., 2013). In brief, the VA ACT-D Training Program incorporates a competency-based training model that includes participation in an experientially-oriented training workshop, followed by weekly, 90minute telephone-based consultation sessions for 6 consecutive months. These sessions are led by an expert ACT-D training consultant who provides feedback on the implementation of the protocol and core processes as well as feedback on and rating of audio-taped therapy sessions. Each clinician trainee needed to complete at least 10 sessions of the ACT-D protocol to meet one of a number of minimum criteria to achieve competency. The ACT-D therapy protocol and training also address assessment and treatment of suicide ideation and risk. Specifically, therapists were instructed to administer the BDI-II prior to each ACT-D session. In addition to examining the BDI-II total score as a possible indicator of suicide risk, therapists also review the BDI-II suicide ideation item (see Beck, Steer, & Brown, 1996), as well as verbally assess for suicidal thought or intent. Any increased or elevated score or verbal report of suicide ideation would then prompt the therapist to evaluate for suicide risk and take appropriate action as clinically indicated (e.g., assess risk level, implement a Safety Plan, etc.) (Wenzel, Brown, & Karlin, 2011). Safety planning consists of a written, prioritized list of coping strategies and sources of support that veterans can use to alleviate a suicidal crisis (Stanley, Brown, Karlin, Kemp, & VonBergen, 2008) and has been used in the context of short-term, evidence-based psychotherapies that have been found to reduce suicide risk (Brown et al., 2005; Stanley et al., 2009). There were no limitations on the types of veteran patients who could receive ACT-D from therapists participating in the training program other than patients needed to have a diagnosis of depression and not be in acute crisis or have an impairment that would render them inappropriate for initiation of psychotherapy. Patients were recruited from the clinician trainees' current practice setting. Patients were either new to the therapist and entered the clinic through regular means (as determined by the clinic) or were already patients of the therapists in the program. No patients were excluded for co-morbidities or other life problems. Patients consented to being audio-taped for the purposes of therapist training and program evaluation. At the first session, patients completed a demographic form that included their age, gender, highest level of education, and ethnicity.

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31

1.2. Measures 1.2.1. Depression The Beck Depression Inventory-II (BDI-II; Beck et al., 1996; Beck, Steer, & Carbin, 1988) was used to assess the severity of depression. The BDI-II consists of 21 items that are scored on a 0e3 scale to reflect the absence or severity of symptoms by patient self-report. The BDI-II total score ranges from 0 to 63, with higher scores indicating greater severity of depression. The BDI-II has been widely used for research as well as clinical purposes and its psychometric properties have been well established (Beck et al. 1988). The BDI-II was administered by therapists before or at the beginning of each therapy session, although only data from the baseline, mid-point (session 7), and final assessment (session 10 or higher) was available to the evaluation team and used in this analysis. 1.2.2. Suicidal ideation The BDI-II also contains an item that consists of 4 statements that describe the severity of suicidal ideation with and without an intent to act on thoughts to kill oneself (Beck et al., 1988).The BDI-II suicide item has been found to be moderately correlated (r's ¼ 0.56e0.58) with the Scale for Suicide Ideation for both inpatient and outpatient psychiatric samples (Beck & Steer, 1991). The predictive validity of this item has been established using data from a prospective study of risk factors for death by suicide in psychiatric outpatients (Brown et al., 2000; Desseilles et al., 2012). 1.2.3. Experiential acceptance The Acceptance and Action Questionnaire-II (AAQ-II; Bond et al., 2011; Hayes, Barnes-Holmes, & Roche, 2001) is a 10-item measure of experiential avoidance that assesses “the degree to which an individual fuses with thoughts, avoids feelings, and is unable to act in the presence of difficult private events, e.g., “My painful memories prevent me from having a fulfilling life” (Hayes et al., 2006). Higher scores indicate greater experiential acceptance and psychological flexibility (i.e., lower experiential avoidance). Items are rated on a Likert scale of 1e7, with several reverse scored items. Bond et al. (2011) evaluated the AAQ-II in a population of 2816 participants and found the mean alpha coefficient to be 0.84, and the 3- and 12-month test-retest reliability to be 0.81 and 0.79, respectively. The alpha for the AAQ-II at baseline in the initial evaluation of the program (Walser et al., 2013) was 0.88 (N ¼ 243). The AAQ-II was administered at baseline, mid-point (session 7), and final assessment (session 10 or higher). 1.2.4. Mindfulness The Five Facet Mindfulness Questionnaire (FFMQ; Baer, Smith, & Allen, 2004; Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006) is a 39-item questionnaire that assesses the respondents' general tendency to be mindful in daily life (e.g., “When I do things, my mind wanders off and I am easily distracted”). The FFMQ has 5 subscales: Observe, Describe, Act with Awareness, Non-judgment, and Nonreaction. It is scored on a 5-point Likert-type scale ranging from 1 (never or very rarely true) to 5 (very often or always true). The FFMQ is based on a factor analytic study including five independently developed mindfulness questionnaires and is considered to have good construct validity (Baer et al., 2008). In a sample of 613 undergraduate students, Baer et al. (2008) found all 5 subscales to have good internal consistency, with the following alphas: Observe ¼ 0.83, Describe ¼ 0.91, Act with Awareness ¼ 0.87, Nonjudgment ¼ 0.87, Non-reaction ¼ 0.75. Additionally, Bohlmeijer, ten Klooster, Fledderus, Veehof, and Baer (2011) found the FFMQ to be a reliable and valid measure in a sample of 376 adults with clinically relevant symptoms of depression and anxiety. For the purposes of mindfulness analyses in this paper, scores from the five subscales

27

were summed to create an overall mindfulness score. This is consistent with a general mindfulness construct (see Baer et al., 2006; Baer et al., 2008). The FFMQ was administered at baseline, mid-point (session 7), and final session (session 10 or higher). 1.3. Statistical analyses The statistical analyses in this paper include descriptive statistics and intent to treat mixed effects models to assess changes in depression severity and suicide ideation over time, during patients' participation in ACT therapy, and to assess the effects of changes in experiential acceptance and mindfulness on changes in depression and SI. The random effects used in the depression severity model allowed the intercepts and slopes of change for each patient nested within their respective therapist to vary, and the random effect used in the suicide ideation model allowed for the intercepts of each patient nested within their respective therapist to vary. All models controlled for age, gender, and race. Furthermore, all continuous covariates were mean centered for increased interpretability of the intercept. The fit of the model was determined to be adequate by visual inspection of the distribution of the standardized error terms. Additionally, the random effects were visually inspected to determine if the assumption of a normal distribution was met. 2. Results 2.1. Patient description There were 981 veteran patients who received ACT-D by therapists who participated in the Training Program. Of these, 741 (75.5%) were men, 222 (22.6%) were women, and 18 (1.8%) did not indicate their gender. The mean patient age was 50.5 years (SD ¼ 12.5) and ranged from 21 to 90 years. Of the 981 patients, 708 (72.2%) were White, 143 (14.6%) were African American/Black, 1 (0.1%) were Asian Indian, 10 (1.0%) were American Indian/Alaskan Native, 4 (0.4%) were Asian, 2 (0.2%) were Pacific Islander, 46 (4.7%) were Multi-Ethnic, and 37 (3.8%) reported themselves as Other. Thirty (3.1%) did not answer the question on race. Out of 981, 176 (17.9%) had an education level less than or equal to high school, 451 (46%) had attended some college, 185 (18.9%) were college graduates, 140 (14.3%) had earned a graduate degree or had attended some graduate school, and 29 (3.0%) did not answer the question on education level. Of the 981 patients, 647 (66.0%) competed 10 or more sessions of ACT or finished early due to symptom relief. The remaining 334 (34.0%) dropped out of therapy prior to completion (n ¼ 177), started too late to finish therapy in time for data analysis (n ¼ 46), died prior to completion of therapy (n ¼ 1), were lost to follow-up because their therapist dropped out of training (n ¼ 37), had a therapist that extended supervision and therefore final data were not available (n ¼ 16), were unable to attend regularly (n ¼ 42), or were lost to follow-up for unknown reasons (n ¼ 15). Of the 934 patients who responded at baseline to the BDI-II item assessing suicidal ideation (SI), 413 (44.2%) had no SI (BDI-II SI item ¼ 0); 472 (50.5%) had SI but no suicidal intent (BDI-II SI item ¼ 1); 37 (4.0%) indicated they would like to kill themselves (BDI-II SI item ¼ 2), and 12 (1.3%) indicated they would do so if they had the chance (BDI-II SI item ¼ 3). Table 1 shows suicidal ideation category at baseline by demographic variable categories and by subsequent ACT-D treatment completion status. A dichotomous variable defined as any SI (BDI-II SI item >0) vs. no SI (BDI-II SI item ¼ 0) was used for subsequent analysis of SI during ACT-D treatment because of low numbers of patients with more severe categories of SI (see Brown et al., 2005; Bruce et al., 2004).

28

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31

Table 1 Suicidal ideation at baseline by demographic variables and by subsequent treatment completion.

Men Women Age <65 Age 65 White Minority race High school or less Some college College grad Some grad school/grad degree Completed ACTa Did not complete ACT a

No suicidal ideation

Suicidal ideation but no intent

Would like to kill themselves

Would do so if had the chance

295 (41.7%) 115 (55.0%) 365 (44.8%) 44 (45.4%) 295 (44.0%) 109 (46.4%) 77 (45.8%) 180 (42.0%) 84 (47.7%) 61 (45.5%) 275 (44.9%) 138 (43.0%)

376 (53.1%) 84 (40.2%) 406 (49.9%) 49 (50.5%) 339 (50.6%) 116 (49.4%) 87 (51.8%) 220 (51.3%) 85 (48.3%) 66 (49.3%) 309 (50.4%) 163 (50.8%)

31 (4.4%) 5 (2.4%) 32 (3.9%) 4 (4.1%) 30 (4.5%) 5 (2.1%) 2 (1.2%) 24 (5.6%) 5 (2.8%) 5 (3.7%) 26 (4.2%) 11 (3.4%)

6 (0.8%) 5 (2.4%) 11 (1.4%) 0 (0.0%) 6 (0.9%) 5 (2.1%) 2 (1.2%) 5 (1.2%) 2 (1.1%) 2 (1.5%) 3 (0.5%) 9 (2.8%)

Completed 10 or more sessions or finished early due to symptom relief.

2.2. Depression severity

2.3. Suicidal ideation

For veterans who completed the BDI-II at baseline and final assessment, those with suicidal ideation, had a mean BDI-II total score of 33.5 at baseline and 22.9 at final assessment (see Table 2 for means and s.d.'s), which represents a 32% mean reduction. For veterans with no suicidal ideation at baseline, mean BDI-II total score decreased from 26.3 at baseline to 15.9 at final assessment, which represents a 40% mean reduction. Two mixed models were used to predict depression severity over time. Model 1 tested the effect of SI at baseline on depression severity scores across assessment points. Model 2 included the additional covariates of experiential acceptance and mindfulness (see Table 2 for means and s.d.'s). Both Model 1 and Model 2 demonstrated higher overall BDI-II scores at baseline associated with baseline SI (b ¼ 8.36, p < .001 and b ¼ 5.62, p < .001, respectively). Both models also demonstrated significant time effects, indicating decreasing depression severity during ACT-D treatment (b ¼ 10.73, p < .001 and b ¼ 10.52, p < .001, respectively), which was similar in patients with and without suicidal ideation at baseline (i.e., not significantly different) in Model 1 (b ¼ 0.61, p ¼ .522). However, after adding experiential acceptance and mindfulness (Model 2), the interaction effect of SI at baseline by time was significant (b ¼ 2.81, p ¼ .001), indicating that patients with SI at baseline achieved significantly greater reduction in depression severity during ACT-D treatment than those with no SI at baseline, when using these two variables as covariates. Of note, both experiential acceptance and mindfulness were associated with significantly lower baseline depression severity (b ¼ 0.44, p < .001 and b ¼ 0.09, p < .001, respectively), controlling for other variables in the model. Furthermore, a significant interaction of time was found with mindfulness, which indicates that the attenuating effect of mindfulness on depression severity increased across time (i.e., greater reduction in depression severity) (b ¼ 0.05, p ¼ .042), but this effect was not found with experiential acceptance (b ¼ 0.05, p ¼ .362). Table 3 displays the fixed effects beta coefficient estimates for Models 1 and 2, respectively.

Among the 548 patients who provided SI data at all three assessment points, 55.5%, 42.0% and 35.0% reported SI at baseline, mid-point (week seven), and final assessment, respectively. Two mixed models were used to predict SI (yes or no). Model 3 showed that there was a significant time effect in the reduction of the odds of SI (log(b) ¼ 1.11, p < .001). Experiential acceptance was associated with significantly lower odds of SI at the baseline assessment, controlling for other variables in the model (log(b) ¼ 0.07, p < .001), but mindfulness was not associated with SI at the baseline assessment (log(b) ¼ 0.01, p ¼ .075). Model 4 included the additional covariates of baseline depression severity and the interaction of both experiential acceptance and mindfulness with time. The effects of time on the odds of SI remained significant after controlling for baseline depression severity (log(b) ¼ 1.15, p < .001), The effect of experiential acceptance remained significant after controlling for baseline depression severity (log(b) ¼ 0.03, p ¼ .016) and the effect of mindfulness on odds of SI remained insignificant. Furthermore, a significant interaction of time was found with experiential acceptance, which indicates that the attenuating effect of experiential acceptance on SI increased across time (e.g., as experiential acceptance increased, odds of SI decreased; log(b) ¼ 0.05, p ¼ .023), but this effect was not found with mindfulness. The effect of baseline depression severity on odds of SI at baseline was in the expected direction (i.e., as depression increases so does the odds of SI), (log(b) ¼ 0.06, p < .001). Table 4 displays the odds ratio estimates for the fixed effects in Models 3 and 4, respectively. 3. Discussion The present evaluation found that approximately half of veterans receiving ACT-D from therapists undergoing ACT-D training reported some degree of suicide ideation at baseline assessment. Overall, ACT-D was effective for patients with and without SI at baseline to final assessment, with un-adjusted mean BDI-II total scores decreasing by 32% and 40% in patients with and without

Table 2 Means and standard deviations of outcome and process variables. Measure

Baseline m(sd) [n]

Midpoint m(sd) [n]

Final m(sd) [n]

BDI-II SI (percent ‘yes’) FFMQ AAQ-II

30.79 (10.667) [981] 55.8% [934] 106.59 (18.948) [883] 30.93 (10.246) [926]

23.71 (13.269) [717] 42.9% [676] 111.23 (20.312) [626] 34.82 (10.784) [665]

19.96 (13.993) [629] 34.7% [585] 118.86 (23.206) [520] 38.92 (11.742) [553]

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31

29

Table 3 Depression severity models. Variable Model 1 Intercept Age Gendera Raceb Baseline SIc Timed Baseline SIc * timed Model 2 Intercept Age Gendera Raceb Baseline SIc Experiential acceptance Mindfulness Timed Baseline SIc * timed Experiential acceptance * timed Mindfulness * timed a b c d

Beta coefficient

SE

DF

t e ratio

p-value

31.03 0.08 0.84 2.30 8.36 10.73 0.61

0.42 0.03 0.78 0.72 0.65 0.48 0.95

895 895 895 895 895 1229 1229

74.66 2.89 1.08 3.17 12.80 22.48 0.64

<0.001 0.004 0.279 0.002 <0.001 <0.001 0.522

30.65 0.03 0.52 0.37 5.62 0.44 0.09 10.52 2.81 0.05 0.05

0.35 0.02 0.62 0.59 0.57 0.03 0.02 0.42 0.87 0.05 0.03

836 836 836 836 836 903 903 903 903 903 903

88.43 1.26 0.83 0.63 9.93 13.83 5.50 24.89 3.22 0.91 2.03

<0.001 0.209 0.405 0.530 <0.001 <0.001 <0.001 <0.001 0.001 0.362 0.042

The reference group is Male. The reference group is Minority. The reference group is Negative for Baseline SI. Time ¼ 0 is the reference to the Baseline Assessment.

Table 4 Suicide ideation models. Variable Model 3 Intercept Age Gendera Raceb Experiential acceptance Mindfulness Timec Model 4 Intercept Age Gendera Raceb Baseline depression severity Experiential acceptance Mindfulness Timec Experiential acceptance * timec Mindfulness * timec a b c

Odds ratio (OR)

95% CI for OR

p-value

1.03 1.02 0.48 1.11 0.94 0.99 0.33

0.82, 1.00, 0.33, 0.79, 0.92, 0.98, 0.25,

1.29 1.03 0.69 1.58 0.95 1.00 0.44

0.783 0.007 <0.001 0.547 <0.001 0.075 <0.001

1.01 1.02 0.48 1.21 1.06 0.97 1.00 0.32 0.96 0.99

0.81, 1.01, 0.33, 0.85, 1.04, 0.95, 0.99, 0.24, 0.92, 0.97,

1.27 1.03 0.69 1.72 1.08 0.99 1.01 0.42 0.99 1.01

0.899 0.004 <0.001 0.288 <0.001 0.016 0.992 <0.001 0.023 0.569

The reference group is Male. The reference group is Minority. Time ¼ 0 is the reference to the Baseline Assessment.

baseline SI, respectively. These observed reductions in depression are consistent with previous research on the effects of treating depression demonstrating resulting reductions in SI (Katalin Szanto, Mulsant, Houck, Dew, & Reynolds, 2003) and compare well with other VA roll outs such as Cognitive Behavioral Therapy for Insomnia (CBT-I; 10% reduction in SI from baseline to follow-up; Trockel, Karlin, Taylor, Brown, & Manber, 2015) and a pilot study of ACT for the treatment of suicidal patients showing significant reductions in SI (Ducasse et al., 2014). In fact the number of patients with no SI increased from 44.5% at baseline to 65% at follow-up. This 20.5% reduction in SI from base-line to follow-up is encouraging and is notably an important finding from this evaluation given the concerted effort to reduce suicide and suicidal ideation in veterans. In reviewing the findings, it might be anticipated that the SI group would have higher depression scores overall and across time.

Accordingly, a differential response to treatment might be expected, with perhaps a slower response by the SI group. This was not the case in the present evaluation. Both groups steadily improved across time, indicating that patients with and without SI at baseline responded to treatment at a similar rate. After adjusting for experiential acceptance and mindfulness, patients with SI at baseline appeared to achieve greater reduction in depression severity during ACT-D treatment than patients with equivalent levels of experiential acceptance and mindfulness but no baseline SI. While research is needed to demonstrate causality, this finding may suggest that the most affected patients (i.e., those with baseline SI) may achieve the greatest reduction in depression severity during ACT-D if they are able to achieve engagement in ACT-D processes that result in experiential acceptance and mindfulness scores equivalent to the scores on these measures achieved by less affected patients (those with no baseline SI). This proposition is also consistent with the finding that the attenuating effect of mindfulness on depression severity became more pronounced across time. Interpreting the same interaction effect in the other direction indicates that greater reductions in depression severity across time were observed in patients with higher mindfulness scores. These overall findings are consistent with other mindfulness approaches that have been found to be effective in treating depression (e.g., Ma & Teasdale, 2004; Teasdale et al., 2000). Experiential acceptance was associated with decreases in suicidal ideation, even after controlling for baseline depression severity. Furthermore, the attenuating effect of experiential acceptance on SI became more pronounced across time. Interpreting the same interaction effect in the other direction indicates greater reductions in odds of SI across time during ACT-D therapy. These effects are novel and are also consistent with the specific process targets of ACT-D as well as research showing that increases in experiential acceptance are correlated with reductions in SI (Ducasse et al., 2014). As part of the ACT intervention, the patient is taught to observe thoughts (e.g., suicidal ideation) and emotions (e.g., sadness and/or pain) as dynamic events that in and of themselves, are not problematic. Rather, it is the response (e.g., escape or avoidance) to these events that can lead to life problems (Luoma & Villatte, 2012). According to the ACT model, excessive and misapplied efforts to control these difficult thoughts and

30

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31

painful emotions may paradoxically increase the patient's difficulty (i.e., suffering about one's pain). These very control efforts may increase the likelihood of the negative consequences created by avoidant behavior. ACT therapy teaches patients to practice experiential acceptance ofdrather than focusing on the escape fromdunwanted thoughts and emotions. This acceptance approach has the potential to disrupt depressive and ruminative processes (Jain et al., 2007) as well as problematic suppression (i.e., avoidance) of negative thoughts (Pettit et al., 2009) by orienting the patient to observation of these experiences, while simultaneously working to make values-based, life-enhancing choices. This disruption in the avoidance process may allow for a more adaptive response to distress and relatedly, less need for harmful escape (i.e., suicide). Research designed to explore and evaluate these hypotheses is needed. There is an accumulation of data that shows that both mindfulness and experiential acceptance may account for the onset and maintenance of various forms of psychological problems (Hayes et al., 2006). They both appear to moderate a variety of relationships between harmful psychological factors and behavioral health outcomes as well (Hayes et al., 2006). Both experiential acceptance and mindfulness processes, as delivered in this program, involved teaching clients to contact the present moment and/or increase willingness to remain present to the ongoing experience of emotion and thought. However, experiential acceptance includes the additional process of taking action in the face of challenging thoughts and emotional experiences, not simply observing them. Clients are not only taught to recognize a negatively evaluated thought as simply a thought (e.g., observe thinking), they are also supported in making and keeping behavioral commitments that are in line with their personal values. These strategies for living with intense emotional experience offer alternatives to avoidance including SI (as conceptualized as avoidance), which may explain the association between experiential acceptance and the observed greater reduction in odds of SI. Further research is needed to test this hypothesis and to further assess the relationship between experiential acceptance and SI. Overall, these findings are in line with other mindfulness-based treatments aimed at reducing suicidal behavior and depression. For instance, both dialectal behavior therapy (DBT; Linehan et al., 2006) and mindfulness -based cognitive therapy (MBCT; Segal, Williams, & Teasdale, 2002) use mindfulness based interventions that mitigate factors contributing to suicidal behavior. The results of the current evaluation are, to our knowledge, the only findings that show reductions in suicide ideation among veterans receiving ACTD in real-world outpatient treatment settings. Although these findings are promising, there are a number of limitations to this evaluation. First, the findings are from patients being treated by therapists who were part of a national training program. Although the therapists were generally quite novice in providing ACT-D, the ongoing consultation may have ensured fidelity to the ACT-D treatment model in ways that may not generalize to ACT-D treatment outside of a training protocol. Second, conclusive statements regarding the efficacy of ACT-D for decreasing the severity of depression and suicide ideation cannot be made given that this was a real-world effectiveness evaluation, rather than a randomized controlled trial. Other factors, such as passage of time may have also had an influence on SI. Furthermore, given the smaller numbers of Veterans who indicated suicidal intent, future research is needed to evaluate whether these findings generalize to such samples. Third, data on other potential confounding variables that may be associated with the reductions in suicidal ideation and depression, such as changes in comorbid medical and mental health conditions or other concurrent treatments, were not available. Finally, as there were no data collected

beyond post-training (follow-up) for the Veterans being treated, we are unable to comment on the long-term outcomes. Future research should include collection of follow-up data to evaluate maintenance of the observed effects of ACT-D on outcome variables related to depression and SI. In summary, veterans receiving ACT-D, on average, achieved significant reductions in depression severity and suicidal ideation during the course of treatment. Furthermore, two key ACT-D intervention processes, increases in mindfulness and increases in experiential acceptance across time, were associated with decreases in depression severity and decreases in SI across time, respectively. These findings, taken together, provide support for the utility and effectiveness of ACT-D in the treatment of depression and suicidal ideation. Further evaluation of the efficacy of ACT-D in reducing suicidal ideation and depression and potential mediators is warranted. References Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of mindfulness by selfreport the Kentucky inventory of mindfulness skills. Assessment, 11(3), 191e206. Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using selfreport assessment methods to explore facets of mindfulness. Assessment, 13, 27e45. Baer, R. A., Smith, G. T., Lykins, E., Button, D., Krietemeyer, J., Sauer, S., et al. (2008). Construct validity of the five facet mindfulness questionnaire in meditating and nonmeditating samples. Assessment, 15(3), 329e342. Beck, A. T., & Steer, R. A. (1991). Manual for the Beck scale for suicide ideation. San Antonio, TX:: Psychological Corporation. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck depression inventory-II. San Antonio, TX: Psychological Corporation. Beck, A. T., Steer, R. A., & Carbin, M. G. (1988). Psychometric properties of the Beck depression Inventory: twenty-five years of evaluation. Clinical Psychology Review, 8(1), 77e100. Bohlmeijer, E., ten Klooster, P. M., Fledderus, M., Veehof, M., & Baer, R. (2011). Psychometric properties of the five facet mindfulness Questionnaire in depressed adults and development of a short form. Assessment, 18(3), 308e320. http://dx.doi.org/10.1177/1073191111408231. Bond, F. W., Hayes, S. C., Baer, R. A., Carpenter, K. M., Guenole, N., Orcutt, H. K., et al. (2011). Preliminary psychometric properties of the acceptance and action QuestionnaireeII: a revised measure of psychological inflexibility and experiential avoidance. Behavior Therapy, 42(4), 676e688. http://dx.doi.org/10.1016/ j.beth.2011.03.007. Brown, G. K., Beck, A. T., Steer, R. A., & Grisham, J. R. (2000). Risk factors for suicide in psychiatric outpatients: a 20-year prospective study. Journal of Consulting and Clinical Psychology, 68(3), 371. Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. X., Hollander, J. E., & Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts. JAMA: the Journal of the American Medical Association, 294(5), 563e570. Bruce, M. L., Ten Have, T. R., Reynolds, C. F., III, Katz, I. I., Schulberg, H. C., Mulsant, B. H., et al. (2004). Reducing suicidal ideation and depressive symptoms in depressed older primary care patients. JAMA: the Journal of the American Medical Association, 291(9), 1081e1091. Chakravorty, S., Grandner, M. A., Mavandadi, S., Perlis, M. L., Sturgis, E. B., & Oslin, D. W. (2013). Suicidal ideation in Veterans misusing alcohol: relationships with insomnia symptoms and sleep duration. Addictive Behaviors, 39(2), 399e405. http://dx.doi.org/10.1016/j.addbeh.2013.09.022. Chiles, J. A., & Strosahl, K. D. (2005). Clinical manual for assessment and treatment of suicidal patients. Arlington, VA US: American Psychiatric Publishing, Inc. Desseilles, M., Perroud, N., Guillaume, S., Jaussent, I., Genty, C., Malafosse, A., et al. (2012). Is it valid to measure suicidal ideation by depression rating scales? Journal of Affective Disorders, 136(3), 398e404. http://dx.doi.org/10.1016/ j.jad.2011.11.013. , E., Be ziat, S., Guillaume, S., Courtet, P., & Olie , E. (2014). AccepDucasse, D., Rene tance And Commitment Therapy for management of suicidal patients: a pilot study. Journal of Psychotherapy and Psychosomatics, 83, 374e376. http:// dx.doi.org/10.1159/000365974. Gotlib, I. H., & Hammen, C. L. (2002). Handbook of depression. New York: Guilford Press. Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational frame theory: A postSkinnerian account of human language and cognition. New York: Plenum Press. Hayes, S. C., Luoma, J. B., Bond, F. W., Masuda, A., & Lillis, J. (2006). Acceptance and commitment Therapy: model, processes and outcomes. Behavior Research and Therapy, 44, 1e25. Hayes, S. C., Strosahl, K., & Wilson, K. G. (2012). Acceptance and commitment therapy. New York: Guilford. Institute of Medicine. (2010). The future of nursing: Leading change, advancing health. Jain, S., Shapiro, S. L., Swanick, S., Roesch, S. C., Mills, P. J., & Schwartz, G. E. (2007). A randomized controlled trial of mindfulness meditation versus relaxation

R.D. Walser et al. / Behaviour Research and Therapy 74 (2015) 25e31 training: effects on distress, positive states of mind, rumination, and distraction. Annals of Behavioral Medicine, 33(1), 11e21. Karlin, B. E., & Cross, G. (2014). From the laboratory to the therapy room: national dissemination and implementation of evidence-based psychotherapies in the U.S. Department of Veterans affairs health care system. American Psychologist, 69, 19e33. Kashdan, T. B., & Rottenberg, J. (2010). Psychological flexibility as a fundamental aspect of health. Clinical Psychology Review, 30, 865e878. http://dx.doi.org/ 10.1016/j.cpr.2010.03.001. Katalin Szanto, K., Mulsant, B. H., Houck, P., Dew, M. A., & Reynolds, C. F., III (2003). Occurrence and course of suicidality during short-term treatment of late-life depression. Archives of General Psychiatry, 60, 610e617. http://dx.doi.org/ 10.1001/archpsyc.60.6.610. Katz, I. (2012). Lessons learned from mental health enhancement and suicide prevention activities in the Veterans health administration. American Journal of Public Health, 102, S14eS16. http://dx.doi.org/10.2105/AJPH.2011.300582. Kuehn, B. M. (2009). Soldier suicide rates continue to rise. JAMA: The Journal of the American Medical Association, 301(11), 1111e1113. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., et al. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Archives of General Psychiatry, 63(7), 757. Luoma, J. B., & Villatte, J. L. (2012). Mindfulness in the treatment of suicidal individuals. Cognitive and Behavioral Practice, 19(2), 265e276. Mann, J. J., Apter, A., Bertolote, J., Beautrais, A., Currier, D., Haas, A., et al. (2005). Suicide prevention strategies. JAMA: the journal of the American Medical Association, 294(16), 2064e2074. Ma, H. S., & Teasdale, J. D. (2004). Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects. Journal of Consulting and Clinical Psychology, 72(1), 31e40. http:// dx.doi.org/10.1037/0022-006X.72.1.31. Pettit, J. W., Temple, S. R., Norton, P. J., Yaroslavsky, I., Grover, K. E., Morgan, S. T., et al. (2009). Thought suppression and suicidal ideation: preliminary evidence in support of a robust association. Depression and Anxiety, 26(8), 758e763. http://dx.doi.org/10.1002/da.20512. € rde Sive Vo €rding, M. B., & Emmelkamp, P. M. (2009). Powers, M. B., Zum Vo Acceptance and commitment therapy: a meta-analytic review. Psychotherapy

31

and Psychosomatics, 78(2), 73e80. http://dx.doi.org/10.1159/000190790. Ruiz, F. J. (2010). A review of Acceptance and Commitment Therapy (ACT) empirical evidence: correlational, experimental psychopathology, component and outcome studies. International Journal of Psychology and Psychological Therapy, 10(1), 125e162. Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-based cognitive therapy for depression: A new approach to preventing relapse. New York: Guilford Press. Stanley, B., Brown, G. K., Brent, D. A., Wells, K., Poling, K., Curry, J., et al. (2009). Cognitive-behavioral therapy for suicide prevention (CBT-SP): treatment model, feasibility, and acceptability. Journal of the American Academy of Child & Adolescent Psychiatry, 48(10), 1005e1013. http://dx.doi.org/10.1097/ CHI.0b013e3181b5dbfe. Stanley, B., Brown, G. K., Karlin, B., Kemp, J. E., & VonBergen, H. A. (2008). Safety plan treatment manual to reduce suicide risk: Veteran version. Washington, DC: United States Department of Veterans Affairs. Teasdale, J. D., Segal, Z. V., Williams, J. M. G., Ridgeway, V. A., Soulsby, J. M., & Lau, M. A. (2000). Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consulting and Clinical Psychology, 68(4), 615e623. http://dx.doi.org/10.1037/0022-006X.68.4.615. Trockel, M., Karlin, B. E., Taylor, C. B., Brown, G. K., & Manber, R. (2015). Effects of cognitive behavioral therapy for insomnia on suicidal ideation in veterans. Sleep, 38, 259e295. http://dx.doi.org/10.5665/sleep.4410. Walser, R. D., Karlin, B. E., Trockel, M., Mazina, B., & Taylor, C. B. (2013). Training in and implementation of Acceptance and Commitment Therapy for depression in the Veterans health administration: therapist and patient outcomes. Behaviour Research and Therapy, 51(9), 555e563. http://dx.doi.org/10.1016/ j.brat.2013.05.009. Walser, R. D., Sears, K., Chartier, M., & Karlin, B. E. (2015). Acceptance and Commitment Therapy for depression in Veterans. Washington, DC: U.S. Department of Veterans Affairs. in press. Wenzel, A., Brown, G. K., & Karlin, B. E. (2011). Cognitive behavioral therapy for depression in veterans and military service members: Therapist manual. Washington, DC: US Department of Veterans Affairs. Zettle, R. (2007). ACT for depression: A clinician's guide to using acceptance & commitment therapy in treating depression. Oakland, CA: New Harbinger Publications.