آزمایش پایلوت رفتار درمانی دیالکتیکی تحت کنترل با و بدون رفتار درمانی دیالکتیکی پروتکل طولانی منطقی مدت برای زنان خودکشی و خودآسیب مبتلا به اختلال شخصیت مرزی و اختلال استرس پس از سانحه

Abstract Objective This study evaluates the efficacy of integrating PTSD treatment into Dialectical Behavior Therapy (DBT) for women with borderline personality disorder, PTSD, and intentional self-injury. Methods Participants were randomized to DBT (n = 9) or DBT with the DBT Prolonged Exposure (DBT PE) protocol (n = 17) and assessed at 4-month intervals during the treatment year and 3-months post-treatment.

.

Results Sample characteristics The sample was an average age of 32.6 years (SD = 12.0, range = 19–55) and was primarily Caucasian (80.8%) followed by biracial (15.4%) and Asian–American (3.8%). A majority of the sample was single, divorced, separated or widowed (84.6%), had less than a college degree (69.2%), and earned $20,000 or less per year (75.0%). Patients in DBT were more likely to be married than those in DBT + DBT PE (44.4% vs. 0%, χ2(1) = 8.9, p < 0.01). There were no other significant between-condition differences on demographic variables. Clinical characteristics of the sample are presented in Table 1. Table 1. Clinical characteristics of the sample at pre-treatment. DBT (n = 7) DBT + DBT PE (n = 19) Total (n = 26) Intentional Self-Injury History Suicide attempts, past year Any suicide attempt 55.6 58.8 57.7 Total suicide attempts (M ± SD) 1.1 ± 1.2 3.1 ± 6.7 2.4 ± 5.5 Non-suicidal self-injury (NSSI), past year Any NSSI 88.9 100.0 96.2 Total NSSI acts (M ± SD) 20.1 ± 29.4 86.2 ± 115.4 63.3 ± 99.1 Trauma History Lifetime Trauma History (M ± SD) Types of lifetime trauma 10.6 ± 5.8 11.8 ± 4.3 11.4 ± 4.8 Age of trauma onset 7.1 ± 5.5 5.8 ± 4.3 6.2 ± 4.7 Index Traumas Childhood sexual abuse 55.5 47.1 50.0 Adult rape 11.1 17.6 15.4 Childhood physical abuse 0.0 17.6 11.5 Intimate partner violence 22.2 5.9 11.5 Threatened with death/serious harm 11.1 5.9 7.7 Sudden death of friend/loved one 0.0 5.9 3.8 Diagnostic Data Current Axis I disorders Any mood disorder 75.0 87.5 83.3 Any anxiety disorder other than PTSD 87.5 87.5 87.5 Any eating disorder 12.5 12.5 12.5 Any substance use disorder 37.5 43.8 41.7 Total no. Axis I disorders (M ± SD) 4.5 ± 2.4 5.2 ± 2.1 5.0 ± 2.2 Current Axis II Disorders Any Axis II disorder other than BPD 50.0 68.8 62.5 Total no. Axis II disorders (M ± SD) 1.9 ± 1.1 2.1 ± 1.0 2.0 ± 1.0 Global Assessment of Functioning (M ± SD) 44.6 ± 3.7 42.3 ± 3.6 43.0 ± 3.7 Psychiatric Treatment History Any inpatient psychiatric admission, past year 44.4 52.9 50.0 Any ER visit for psychological reasons, past year 77.8 70.6 73.1 Any psychotropic medication, past year 87.5 88.2 88.0 Note: Data are given as percentages unless otherwise indicated. BPD = borderline personality disorder. Table options Treatment feasibility Treatment retention Completion rates for the one year of treatment did not differ between conditions (DBT = 55.6%, DBT + DBT PE = 58.8%; χ2(1) = 0.3, p = 0.87), but were lower than those typically found in DBT studies (73%; Kliem, Kroger, & Kosfelder, 2010). The lower than average completion rate is accounted for by one therapist who was not adherent to DBT and had a 100% dropout rate. Excluding this therapist's four patients, the treatment completion rates were: DBT = 71.4%, DBT + DBT PE = 66.7%. In general, patients could choose reassignment to another therapist, and the number of therapists dropped was significantly correlated with lower therapist adherence to DBT (r = −0.5, p < 0.02). Premature dropout occurred on average at week 19.3 in DBT (SD = 9.7) and 29.3 in DBT + DBT PE (SD = 13.4). Reasons for premature dropout from DBT were: committed suicide (n = 1), not motivated to attend treatment (n = 1), and unknown (n = 2). Reasons for premature dropout from DBT + DBT PE were: time problems due to full-time work and/or school and felt problems had improved (n = 2), out of town for four consecutive weeks (n = 1), practical problems attending treatment due to move (n = 1), and unknown (n = 3). Treatment attendance There were moderate, but non-significant effects indicating that patients in DBT + DBT PE attended a greater number of individual therapy sessions than those in DBT in the ITT sample (DBT: M = 28.8, SD = 16.7, DBT + DBT PE: M = 38.6, SD = 18.5, t(24) = 1.3, p = 0.19, g = 0.5) and among treatment completers (DBT: M = 41.0, SD = 9.3, DBT + DBT PE: M = 48.5, SD = 11.3, t(9) = 1.2, p = 0.27, g = 0.7). The number of skills training groups attended did not differ by condition in the ITT sample (DBT: M = 24.4, SD = 17.8, DBT + DBT PE: M = 28.7, SD = 14.4, t(24) = 0.7, p = 0.51, g = 0.3) or among treatment completers (DBT: M = 38.4, SD = 7.9, DBT + DBT PE: M = 35.8, SD = 5.3, t(9) = 0.6, p = 0.56, g = 0.4). DBT PE protocol implementation Of the 17 patients in the DBT + DBT PE condition, 8 (47.1%) started the DBT PE protocol and 9 (52.9%) did not. Of the 9 patients that did not start the DBT PE protocol, reasons for not starting included: dropped out of treatment (n = 4; 44.4%), PTSD remitted without targeted treatment (n = 3; 33.3%), and did not achieve sufficient stability (n = 2; 22.2%). The 8 patients that started the DBT PE protocol did so at week 21.9 of DBT on average (SD = 11.6, range = 6–37). Of these, 6 (75%) completed the protocol in an average of 12.7 sessions (SD = 2.9, range = 9–17) during which an average of 3.0 trauma memories was targeted (SD = 1.7, range = 1–6). Reasons for non-completion included difficulty controlling NSSI (n = 1) and unwillingness to continue (n = 1). Among patients who started the DBT PE protocol, 2 (25%) were not taking an SSRI and 6 (75%) remained at a stable dose of an SSRI during this portion of the treatment. Treatment acceptability At intake, the majority of patients indicated a preference for a combined DBT and PE treatment (n = 19, 73.1%) and the remainder preferred to receive DBT only (n = 7, 26.9%). No patients reported a preference for PE alone. Treatment preference did not differ between conditions (Fisher's exact p = 0.19). Patients in both conditions reported comparably positive treatment expectancies (DBT: M = 5.6, SD = 1.5, DBT + DBT PE: M = 5.9, SD = 1.0, t(127) = 1.1, p = 0.29, g = 0.2) and were highly satisfied with the treatment they received at post-treatment (DBT: M = 25.3, SD = 7.5, DBT + DBT PE: M = 25.4, SD = 6.4, t(17) = 0.04, p = 0.97, g = 0.02). In addition, therapist treatment expectancies were very positive and did not differ between conditions (DBT: M = 5.8, SD = 0.9, DBT + DBT PE: M = 5.9, SD = 0.8, t(82) = 0.6, p = 0.55, g = 0.1). Treatment safety Pre- and post-session urges to commit suicide and self-injure were higher and more stable among patients in DBT compared to those in DBT + DBT PE (Table 2). Specifically, patients in DBT reported significantly higher pre- and post-session urges to commit suicide (M's = 1.8–2.0) and self-injure (M's = 1.8–2.0) than those in DBT + DBT PE (M's = 1.0–1.3 and 1.1–1.4, respectively). Moreover, there was a significant difference in the pattern of pre–post-session change in urges to commit suicide by condition, which was accounted for by the fact that suicide urges were more likely to decrease after sessions in DBT + DBT PE (29.7% vs. 18.2%; χ2(1) = 8.0, p < 0.01) and to remain unchanged after sessions in DBT (76.7% vs. 60.3%, χ2(1) = 14.1, p < 0.001). The pattern of change in pre–post-session urges to self-injure did not significantly differ between conditions. Table 2. Urges to commit suicide and self-injure before and after individual therapy sessions and exposure tasks. DBT + DBT PE DBT Between-group test Exposure tasks (n = 256) DBT sessions (n = 429) DBT PE sessions (n = 70) Within-group test DBT sessions (n = 159) Urges to Commit Suicide Intensity of Urges, M ± SD Pre 0.3 ± 0.7 1.4 ± 1.6 1.1 ± 1.4 t(560) = 1.5 2.0 ± 2.2 t(736) = 4.2∗∗∗ Post 0.3 ± 0.9 1.1 ± 1.5 0.9 ± 1.3 t(500) = 1.1 1.8 ± 2.0 t(659) = 5.0∗∗∗ Change in Urges, n (%) Decrease in urges 24 (9.4%) 133 (31.0%) 15 (21.4%) χ2(2) = 3.3 29 (18.2%) χ2(2) = 14.3∗∗ No change in urges 211 (82.4%) 252 (58.7%) 49 (70.0%) 122 (76.7%) Increase in urges 21 (8.2%) 44 (10.3%) 6 (8.6%) 8 (5.0%) Urges to Self-Injure Intensity of Urges, M ± SD Pre 0.4 ± 1.0 1.4 ± 1.7 1.1 ± 1.5 t(564) = 1.7 2.0 ± 2.1 t(740) = 4.0∗∗∗ Post 0.5 ± 1.2 1.1 ± 1.6 0.9 ± 1.3 t(501) = 1.0 1.8 ± 2.0 t(660) = 4.7∗∗∗ Change in Urges, n (%) Decrease in urges 28 (10.9%) 141 (32.6%) 16 (22.9%) χ2(2) = 4.5 41 (25.8%) χ2(2) = 4.5 No change in urges 199 (77.7%) 232 (53.6%) 47 (67.1%) 103 (64.8%) Increase in urges 29 (11.3%) 60 (13.9%) 7 (10.0%) 15 (9.4%) Note: Urges were rated on a 0–5 scale. Within-group tests compared DBT to DBT PE sessions within the DBT + DBT PE condition. Between-group tests compared the two conditions across all sessions. ∗∗p < 0.01. ∗∗∗p < 0.001. Table options Among patients in DBT + DBT PE, neither the average intensity nor the pattern of change of pre–post-session urges to commit suicide and self-injure differed between DBT and DBT PE protocol sessions (Table 2). Similarly, urges to commit suicide and self-injure were low both before and after completing exposure tasks (M's = 0.3–0.5) and completing an exposure task rarely led to an increase in urges (<12% of tasks). Of the 8 patients who started the DBT PE protocol, 2 (25.0%) reported a relapse of intentional self-injury during this portion of the treatment (suicide attempt (n = 1), NSSI (n = 1)). Clinical outcomes For all outcomes, descriptive data are shown in Table 3, within- and between-group Hedge's g effect sizes are shown in Table 4, rates of reliable and clinically significant change are shown in Table 5, and results of mixed-effects models are shown in Table 6. Table 3. Descriptive statistics for outcome variables by condition and treatment completer status. Outcome DBT + DBT PE DBT Intent to treat Treatment completers Intent to treat Treatment completers M (SD) Pre (n = 17) Post (n = 12) FU (n = 12) Pre (n = 6) Post (n = 5) FU (n = 5) Pre (n = 9) Post (n = 6) FU (n = 6) Pre (n = 5) Post (n = 5) FU (n = 5) PTSD 32.8 (8.0) 13.6 (13.2) 16.7 (14.1) 30.7 (6.8) 6.2 (8.2) 11.0 (14.3) 30.1 (9.6) 13.8 (9.3) 18.4 (7.7) 30.8 (10.7) 13.8 (10.4) 19.8 (8.2) Suicide attemptsa 1.0 (2.2) 0.0 (0.0) 0.1 (0.3) 0.2 (0.4) 0.0 (0.0) 0.2 (0.5) 0.4 (0.4) 0.2 (0.4) 0.0 (0.0) 0.2 (0.3) 0.2 (0.5) 0.0 (0.0) NSSI actsa 28.8 (38.5) 0.9 (2.0) 0.6 (1.2) 42.0 (49.0) 0.2 (0.5) 0.8 (1.8) 6.7 (9.8) 1.0 (2.0) 0.3 (0.5) 6.3 (7.5) 1.2 (2.2) 0.4 (0.5) Dissociation 22.7 (21.6) 10.0 (16.9) 11.3 (16.7) 23.1 (22.3) 5.8 (3.3) 6.8 (8.7) 21.8 (21.7) 14.6 (12.5) 17.3 (15.3) 21.8 (28.5) 13.8 (13.8) 16.5 (16.9) Trauma-related guilt cognitions 2.4 (0.8) 1.6 (0.8) 1.7 (0.9) 2.7 (0.6) 1.0 (0.6) 1.2 (0.8) 2.4 (0.9) 2.5 (1.0) 2.4 (1.3) 2.8 (0.5) 2.4 (1.1) 2.2 (1.3) Shame 87.7 (10.5) 61.8 (16.1) 65.3 (19.6) 90.7 (7.1) 49.6 (10.6) 59.0 (18.8) 84.1 (13.7) 67.7 (15.3) 66.0 (15.2) 84.0 (14.8) 69.4 (16.5) 67.6 (16.5) Anxiety 25.8 (9.0) 14.2 (10.8) 15.0 (10.6) 27.3 (8.9) 8.4 (6.9) 11.2 (13.6) 27.6 (10.9) 17.8 (8.6) 16.3 (7.0) 25.8 (11.8) 16.8 (9.2) 15.8 (7.7) Depression 22.9 (5.7) 11.8 (8.0) 12.5 (8.2) 21.2 (3.9) 8.2 (4.9) 9.4 (8.9) 25.6 (6.2) 15.5 (6.5) 16.8 (3.4) 25.0 (6.2) 15.0 (7.2) 17.4 (3.5) Global severity 2.6 (0.6) 1.1 (0.7) 1.4 (0.9) 2.6 (0.7) 0.6 (0.4) 1.1 (1.0) 2.2 (0.7) 1.2 (0.5) 1.7 (0.8) 2.3 (1.0) 1.3 (0.5) 1.7 (0.8) Note. In DBT + DBT PE, treatment completion is defined as completing the DBT PE protocol. In DBT, treatment completion is defined as completing one year of DBT. Means and standard deviations were calculated using the observed data. NSSI = non-suicidal self-injury. FU = 3-month follow-up. a Includes the number of episodes in the past 4 months. Table options Table 4. Within- and between-group Hedge's g effect sizes. Outcome Within-group effect sizes Between-group effect sizes DBT + DBT PE DBT DBT + DBT PE vs. DBT Pre-post Pre-FU Pre-post Pre-FU Post FU ITT TC ITT TC ITT TC ITT TC ITT TC ITT TC PTSD 1.8 2.9 1.4 1.6 1.3 1.5 0.9 0.9 0.0 0.7 0.1 0.6 Suicide attempts 0.6 0.8 0.5 0.1 0.4 0.0 1.0 0.9 0.7 0.6 −0.3 −0.6 Non-suicidal self-injury acts 1.0 1.1 1.1 1.1 0.8 0.8 0.9 1.0 0.0 0.6 −0.2 −0.3 Dissociation 0.5 0.8 0.4 0.6 0.5 0.3 0.3 0.2 0.3 0.7 0.4 0.7 Trauma-related guilt cognitions 0.9 2.3 0.8 1.8 0.4 0.5 0.4 0.6 1.0 1.4 0.7 0.8 Shame 1.9 3.9 1.4 2.0 1.1 0.8 1.2 1.0 0.4 1.3 0.0 0.4 Anxiety 1.1 2.0 1.1 1.2 0.7 0.8 0.9 0.9 0.3 0.9 0.1 0.4 Depression 1.7 3.0 1.6 1.7 1.4 1.4 1.6 1.4 0.5 1.0 0.6 1.1 Global severity 2.1 3.3 1.4 1.6 1.3 1.2 0.6 0.5 0.2 1.4 0.2 0.6 Note. Within-group effect sizes were calculated using observed data from those patients who completed both of the relevant assessment points: DBT + DBT PE (ITT n = 12, TC n = 5), DBT (ITT n = 6, TC n = 5). Positive within-group effect sizes indicate improvements in outcomes, whereas negative within-group effect sizes indicate worsening in outcomes. Positive between-group effect sizes indicate a greater improvement in DBT + DBT PE than in DBT. ITT = Intent to treat, TC = Treatment completers, FU = 3-month follow-up. Table options Table 5. Reliable and clinically significant improvement. Intent to treat Treatment completers Reliable improvement Normal functioning Both criteria Reliable improvement Normal functioning Both criteria DBT + DBT PE DBT DBT + DBT PE DBT DBT + DBT PE DBT DBT + DBT PE DBT DBT + DBT PE DBT DBT + DBT PE DBT PTSD Post-treatment 10 (83.3) 4 (66.7) 7 (58.3) 5 (83.3) 7 (58.3) 3 (50.0) 5 (100.0) 4 (80.0) 4 (80.0) 4 (80.0) 4 (80.0) 3 (60.0) Follow-up 8 (66.7) 2 (40.0) 7 (58.3) 1 (20.0) 7 (58.3) 0 (0.0) 4 (80.0) 2 (50.0) 4 (80.0) 0 (0.0) 4 (80.0) 0 (0.0) Trauma-related guilt cognitions Post-treatment 4 (33.3) 1 (16.7) 5 (45.5) 1 (16.7) 3 (27.3) 1 (16.7) 3 (60.0) 1 (20.0) 4 (80.0) 1 (20.0) 3 (60.0) 1 (20.0) Follow-up 3 (25.0) 1 (16.7) 4 (36.4) 1 (16.7) 2 (18.2) 1 (16.7) 3 (60.0) 1 (20.0) 3 (60.0) 1 (20.0) 2 (40.0) 1 (20.0) Shame Post-treatment 9 (75.0) 4 (66.7) 7 (63.6) 3(50.0) 7 (63.6) 2 (33.3) 5 (100.0) 3 (60.0) 5 (100.0) 2 (40.0) 5 (100.0) 1 (20.0) Follow-up 7 (58.3) 3 (50.0) 6 (54.5) 4 (66.7) 6 (54.5) 3 (50.0) 3 (60.0) 2 (40.0) 3 (60.0) 3 (60.0) 3 (60.0) 2 (40.0) Anxiety Post-treatment 7 (58.3) 4 (66.7) 3 (25.0) 0 (0.0) 3 (25.0) 0 (0.0) 4 (80.0) 4 (80.0) 2 (40.0) 0 (0.0) 2 (40.0) 0 (0.0) Follow-up 7 (58.3) 3 (50.0) 3 (25.0) 1 (16.7) 3 (25.0) 1 (16.7) 4 (80.0) 3 (60.0) 3 (60.0) 1 (20.0) 3 (60.0) 1 (20.0) Depression Post-treatment 7 (58.3) 4 (66.7) 5 (41.7) 1 (16.7) 5 (41.7) 1 (16.7) 4 (80.0) 4 (80.0) 3 (60.0) 1 (20.0) 3 (60.0) 1 (20.0) Follow-up 7 (58.3) 4 (66.7) 3 (25.0) 0 (0.0) 3 (25.0) 0 (0.0) 4 (80.0) 3 (60.0) 2 (40.0) 0 (0.0) 2 (40.0) 0 (0.0) Global severity Post-treatment 9 (75.0) 3 (50.0) 5 (41.7) 1 (16.7) 5 (41.7) 0 (0.0) 5 (100.0) 2 (40.0) 4 (80.0) 1 (20.0) 4 (80.0) 0 (0.0) Follow-up 8 (66.7) 3 (50.0) 3 (25.0) 0 (0.0) 3 (25.0) 0 (0.0) 4 (80.0) 2 (40.0) 2 (40.0) 0 (0.0) 2 (40.0) 0 (0.0) Note. All data are presented as n (%). Only patients who had pre-treatment scores larger than the cut-offs for reliable change and clinically significant change were included in the calculations. Table options Table 6. Results of mixed-effects models. Main effects Interactions Time Condition Completer Time × condition Time × Completer Condition × Completer Time × condition × Completer PTSDa 4.53,42∗∗ 0.41,22 0.51,21 0.33,42 2.33,42ˆ 0.61,22 3.43,42∗ Dissociationa 2.83,41∗ 0.51,20 0.31,20 0.53,41 0.23,41 1.11,20 0.13,41 Trauma-related guilt cognitions 0.01,19 0.41,87 9.01,87∗∗ 1.51,19 6.21,19∗ 3.91,87∗ 0.61,19 Shame 21.51,29∗∗∗ 0.11,92 2.01,92 0.01,29 0.31,29 0.11,92 1.61,29 Anxiety 16.51,22∗∗ 0.71,22 0.11,22 0.61,22 2.41,22 0.41,22 0.31,22 Depression 11.91,28∗∗ 2.51,93 3.21,93ˆ 0.01,28 0.11,28 0.11,93 0.11,28 Global severity 17.91,16∗∗ 0.41,23 0.21,23 0.61,16 0.31,16 0.21,23 0.71,16 Note. All results are presented as F-values (Fdf1,df2). Unless otherwise specified, hierarchical linear models were used. ˆp < 0.10. ∗p < 0.05. ∗∗p < 0.01. ∗∗∗p < 0.001. a Mixed model analyses of variance were used to account for the non-linearity of the data. Table options Primary outcomes Intentional self-injury One patient in DBT committed suicide during the study. In the ITT sample, the rate of any suicide attempt during the study was 37.5% in DBT + DBT PE and 50.0% in DBT. Among treatment completers, the rate of any suicide attempt was 16.7% in DBT + DBT PE and 40.0% in DBT. Pre–post changes in the frequency of suicide attempts were moderate to large in DBT + DBT PE (g's = 0.6–0.8) and small in DBT (g's = 0.0–0.4). At post-treatment, between-group effect sizes were moderate in favor of DBT + DBT PE (g's = 0.6–0.7). During the follow-up period, 91.7% of patients in DBT + DBT PE and 100% of patients in DBT were abstinent from suicidal behavior. A similar pattern of results was found for NSSI. In the ITT sample, the rate of any NSSI during the study was 68.8% in DBT + DBT PE and 87.5% in DBT. Among treatment completers, the rate of any NSSI was 66.7% in DBT + DBT PE and 100% in DBT. Pre–post changes in the frequency of NSSI were very large in DBT + DBT PE (g's = 1.0–1.1) and large in DBT (g's = 0.8). At post-treatment, between-group effect sizes were small in the ITT sample (g = 0.0) and moderate in favor of DBT + DBT PE among treatment completers (g = 0.6). During the follow-up period, 75.0% of patients in DBT + DBT PE and 66.7% of patients in DBT were abstinent from NSSI. Among treatment completers, rates of abstinence from NSSI during follow-up were 80.0% in DBT + DBT PE and 60.0% in DBT. Mixed-effects models did not converge for either suicide attempts or NSSI due to the high model complexity and low within-subject variability. PTSD There were very large pre–post changes in PTSD severity in both conditions (g's > 1.2), with the largest effect found for DBT PE protocol completers (g = 2.9). At 3-month follow-up, changes in PTSD severity remained very large in DBT + DBT PE (g's = 1.4–1.6) and large in DBT (g's = 0.9). Between-group effect sizes at post-treatment and follow-up were small in the ITT sample (g's = 0.0–0.1) and moderate among treatment completers (g's = 0.6–0.7). At post-treatment, a majority of patients in the ITT sample had reliably improved and reached a normal level of functioning (i.e., recovered; DBT + DBT PE = 58.3%, DBT = 50.0%), and recovery was highest among DBT + DBT PE treatment completers (80%). At 3-month follow-up, the rate of recovery remained unchanged in DBT + DBT PE, but decreased to 0% in DBT. When using the most stringent criterion of diagnostic remission, this pattern of findings was even more pronounced. At post-treatment, rates of PTSD remission were: DBT + DBT PE (ITT = 58.3%, completers = 80.0%) and DBT (ITT = 33.3%, completers = 40.0%). At 3-month follow-up, no patients in DBT (0%) were in remission, whereas remission rates remained high in DBT + DBT PE (ITT = 50.0%, completers = 60.0%). Finally, a MMANOVA found a significant reduction in PTSD severity across time. However, a significant Time × Condition × Completer interaction indicated that completers in DBT + DBT PE showed the largest improvement in PTSD severity. Secondary outcomes Within-group effect sizes for dissociation, trauma-related guilt cognitions, shame, anxiety, depression, and global severity were generally very large in DBT + DBT PE (ITT average g = 1.2, range = 0.4–2.1), particularly among DBT PE protocol completers (average g = 2.0, range = 0.6–3.9). In DBT, within-group effect sizes on these secondary outcomes were generally large in both the ITT sample (average g = 0.9, range = 0.3–1.6) and among treatment completers (average g = 0.8, range = 0.2–1.4). At post-treatment, between-group effect sizes in favor of DBT + DBT PE were generally moderate in the ITT sample (average g = 0.5, range = 0.2–1.0) and very large among treatment completers (average g = 1.1, range = 0.7–1.4). At 3-month follow-up, between-group effect sizes were generally small in the ITT sample (average g = 0.3, range = 0.0–0.7) and moderate among treatment completers (average g = 0.7, range = 0.4–1.1). In addition, 60–100% of treatment completers in DBT + DBT PE both reliably improved and reached normal levels of functioning (i.e., recovered) for all secondary outcomes at post-treatment and/or follow-up. In contrast, 0–20% of treatment completers in DBT recovered on each secondary outcome at post-treatment and follow-up with the exception of shame at follow-up (40%). Mixed-effects models found significant reductions across time for all secondary outcomes except trauma-related guilt cognitions. However, significant Time × Completer and Condition × Completer interactions were found for trauma-related guilt cognitions, indicating that completers showed a significantly greater improvement than non-completers across time, and this was particularly true for completers in DBT + DBT PE. In addition, simple slopes analyses indicated that anxiety decreased significantly in DBT + DBT PE (β = −3.0, t(54) = 4.6, p < 0.0001), but not in DBT (β = −2.0, t(54) = 1.9, p = 0.06). Similarly, there was a significant reduction in global severity in DBT + DBT PE (β = −0.3, t(54) = 5.1, p < 0.0001), but only a trend in DBT (β = −0.2, t(54) = 2.0, p = 0.05).