Trauma-Integrated Mindfulness-Based Relapse Prevention for Women with Comorbid Post-Traumatic Stress Disorder and Substance Use Disorder: A Cluster Randomized Controlled Feasibility and Acceptability Trial

Abstract

Background:

Comorbid post-traumatic stress disorder and substance use disorder (PTSD-SUD) among women receiving substance use treatment are common. Few evidence-based interventions target PTSD-SUD, however, fewer are gender responsive. Mindfulness-based relapse prevention (MBRP) has shown effectiveness for women with SUD, although it does not explicitly target PTSD. Integration of trauma-focused and gender-responsive treatments into MBRP may address the limited availability of PTSD-SUD interventions for women. This study assessed feasibility and acceptability of trauma-integrated MBRP (TI-MBRP).

Methods:

A single-blind computer-generated cluster-randomized design was employed in which women with PTSD-SUD (N = 83) received either TI-MBRP (k = 5) or MBRP (k = 5). Measures of PTSD symptom severity and craving were administered at pre-, post-, 1-, 3-, 6-, 9-, and 12-month follow-up and assessed at the individual level.

Results:

TI-MBRP demonstrated acceptability among participants; however, attrition was high (64%) at 12-month follow-up. Reductions in PTSD were greater in the MBRP than in the TI-MBRP group at postcourse and 1-month follow-up, and there were significant reductions in PTSD severity and craving over the 12-month period in both conditions.

Conclusions:

Integrating trauma- and gender-focused interventions into MBRP was feasible and acceptable. MBRP alone may be effective in reducing both PTSD and SUD symptoms in women with PTSD-SUD; however, confirmatory studies are warranted.

Clinical Trial Registration Number:

NCT03505749.

Introduction

Post-traumatic stress disorder (PTSD) among women in substance use disorder (SUD) treatment is common.¹ Women with comorbid PTSD-SUD are more likely to use substances to cope with trauma-related affect^2–5 than men with PTSD-SUD.⁶ Compared with women with either disorder alone, those with PTSD-SUD are more likely to use “harder” drugs (e.g., cocaine, opiates), have more severe psychiatric conditions, poorer treatment outcomes, and greater psychosocial deficits.^7–11 Given these heightened risks and consequences, and studies suggesting PTSD should be treated in conjunction with SUD^9,12,13 versus treating each separately,^13–15 investigating evidence-based interventions (EBIs) targeting PTSD-SUD is imperative.

Although some treatment models target PTSD-SUD concurrently,^16–19 availability of gender-responsive EBIs for women with PTSD-SUD is scarce and needed.^20–25 The role of substance use within interpersonal relationships and co-occurring mental health issues may be more central to women's SUD treatment than to men's.²⁶ Furthermore, EBIs adapted to be trauma-informed, women-specific/women-only, and inclusive of gender-focused topics demonstrate improved treatment completion, SUD and psychiatric symptoms, and functional outcomes.^27–30 Only two locatable EBIs for PTSD-SUD have been validated on dually diagnosed women,^31,32 neither of which integrate gender-responsive approaches.

Mindfulness-based interventions have shown clinical effectiveness for SUD,³³ and more recently for individuals with PTSD.^34,35 MBIs demonstrate positive effects on mechanisms underlying both PTSD and SUD.^35–41 Moreover, women with SUD have shown preference for MBIs,⁴² and for integrated PTSD-SUD treatments versus SUD-only interventions.⁴³

Mindfulness-based relapse prevention (MBRP),⁴⁴ a widely disseminated EBI, has shown efficacy in reducing SUD outcomes among an array of substances and within trauma-exposed populations.^45–48 MBRP uses an 8-week standardized protocol that integrates mindfulness practice, cognitive–behavioral approaches, and psychoeducation to increase awareness of habitual patterns associated with addictive behaviors and factors precipitating and maintaining substance use. It focuses on skillful responses in high-risk situations, and integrating behaviors that support lifestyle wellness.³³

Although trauma-processing techniques are not part of standard MBRP, evidence suggests MBRP may be suitable for women with PTSD-SUD.⁴⁸ Integrating relevant gender- and trauma-focused techniques into standard MBRP may provide a more robust approach for dually diagnosed women. Only two locatable studies have adapted and tested MBIs for PTSD-SUD populations: one adapting mindfulness-based stress reduction for women in SUD treatment⁴⁹ and the other integrating mindfulness meditation with cognitive therapy for adolescents.⁵⁰ Given the enduring prevalence of PTSD-SUD, and the unique needs and risks women with this comorbidity face, further development and assessment of effective integrated treatment are warranted.

Current study

This study evaluated an adapted program integrating cognitive processing therapy (CPT),³² an EBI for PTSD, and gender-focused topics into standard MBRP for women with PTSD-SUD.

The study was designed to (1) determine feasibility and acceptability of trauma-integrated MBRP (TI-MBRP) as measured by recruitment, retention, homework compliance, and course satisfaction and (2) conduct exploratory analyses to assess changes in target outcomes (i.e., craving and PTSD symptom severity) over a 12-month follow-up period. Since participants underwent treatment programming within established cohorts, they were randomized by cluster (i.e., treatment cohort) to either TI-MBRP or MBRP. All outcomes were assessed at the individual level.

All procedures were approved by the associated institutional IRB (CTR No. NCT03505749). Analyses of within-subjects change described in the original registered protocol were excluded due to small sample size. Furthermore, candidate moderators and mediators listed in the original protocol were excluded from this study due to insufficient power to assess mechanisms over time.

Materials and Methods

Trauma-integrated mindfulness-based relapse prevention

Informed by patients who formerly underwent standard MBRP and clinical staff, an eight-session protocol integrating CPT and gender-focused targets (e.g., interpersonal relationships, parenting, empowerment)^51,52 was developed. CPT is designed to bring awareness to trauma-related beliefs often centering on exaggerated blame toward the trauma survivor or overgeneralization of trauma cues to objectively safe situations, which maintain PTSD symptoms.

Restructuring beliefs to be more realistic reduces trauma-related affect and cognitions, and facilitates trauma processing.⁵³ Each TI-MBRP session included mindfulness practices designed to bring awareness to cognitive and behavioral processes underlying substance use, reduce initial reactivity and promote nonjudgmental acceptance, and explore substance use as a mechanism to cope with PTSD symptoms.

Selection of participants and sampling

Women in the stabilization phase of treatment programming were recruited from three residential and two intensive outpatient (IOP) community-based SUD treatment centers in the Pacific northwest region of the United States. Criteria for admittance into participating facilities included diagnosis, by a licensed mental health provider, of SUD and another mental health disorder. Approximately 95% of women in these facilities had a lifetime history of trauma, and ∼88% were diagnosed with PTSD. All women were required to take psychoeducational and process-focused courses on substance use, domestic violence, and parenting. None of the clinics offered integrated evidence-based PTSD-SUD courses. Duration of both inpatient and IOP treatment programs was ∼90 days.

Clinic staff assisted with recruitment through word of mouth, flyers, and announcements at weekly meetings. Additional efforts were made to recruit ethnic and racial minority participants to ensure the study sample represented the local population. Inclusion criteria were (1) having undergone detoxification from substance use, (2) a score of at least four on Breslau's Short Screening Scale (BSSS),⁵⁴ (3) English fluency, (4) clearance from clinical staff, and (5) being between ages 18 and 70 years. Individuals were excluded if they (1) endorsed active suicidality or hallucinations, (2) previously participated in MBRP, or (3) did not provide informed consent. All assessments were administered at the clinical site from which participants were recruited.

Sample size estimation

For a fully powered cluster-randomized design, a power analysis,⁵⁵ using α = 0.05, medium effect size across primary outcomes, 2.55 variance inflation factor, and power level of 0.8, suggests a sample size of 437 participants.⁵⁶ As a feasibility trial, it is justified to recruit 10% of that required for a fully powered trial^57,58; thus, we aimed to recruit 100 participants powered for 1-month outcomes, estimating 56% attrition.⁴⁸ Although anticipated to be underpowered, 12-month follow-up was included to assess evidence-based methods for retaining hard-to-reach clinical research populations.^59,60

Measures

Screening and participant demographics

The BSSS⁵⁴ is a 7-item self-administered screening used to detect PTSD symptoms in primary care settings. A score of four or more indicates likely PTSD diagnosis. Primary clinicians confirmed diagnoses for eligible participants. A brief demographic survey collected information on sociodemographic variables.

Feasibility and acceptability

Feasibility was evaluated by recruitment, enrollment, and retention.⁶¹ Recruitment was assessed by number of participants who completed informed consent and screening, with at least 30% identifying as racial and/or ethnic minority. Enrollment was measured by percentage of eligible individuals who completed baseline measures. Based on retention in a randomized-controlled trial conducted at a similar clinic,⁴⁸ it was aimed for at least 60% retention at 1-month follow-up.

Acceptability was assessed by homework compliance and course satisfaction.⁶² Homework compliance was measured by self-reported duration of formal (target of at least 70 min per week) and frequency of informal (target of 7 instances per week) mindfulness practice. Course satisfaction was assessed using the Overall Course Satisfaction Survey (OCSS),^44,46 a 10-point Likert-type scale rating importance of MBRP to SUD recovery, and likelihood of treatment enactment (i.e., continued mindfulness practice) after the course.

Preliminary outcomes

The Penn Alcohol Craving Scale (PACS)⁶³—adapted to assess craving for any substance, a 5-item self-report measure—assessed frequency, intensity, and duration of alcohol and/or drug craving over the prior week. Items are scaled from 0 (“Never”) to 6 (“All the time”). Internal consistency for this study was excellent (α = 0.93). The post-traumatic checklist for the DSM-5 (PCL-5), a 20-item self-report measure assessed PTSD symptoms over the past week.⁶⁴ The scale is rated from 0 (“Not at all”) to 4 (“Extremely”) for each symptom. Internal consistency for this study was excellent (α = 0.91).

Procedures

Research staff provided study information and oral and written consent at each site, and screened interested individuals for eligibility. Eligible individuals completed baseline assessment within 1 week before the start of the intervention. Single-blind randomization through a computer generator allocated 10 clusters of participants to eight 90-min sessions of either standard MBRP or TI-MBRP over the course of 4 weeks. To avoid cross-contamination, only one condition was offered at each site at a time. A total of five TI-MBRP and five MBRP groups were conducted over 18 months. Graduate-level research assistants facilitated intervention groups; two were dedicated MBRP interventionists and two were dedicated TI-MBRP interventionists.

All interventionists underwent formal MBRP training, and TI-MBRP interventionists underwent additional training to implement trauma-processing components. MBRP interventionists did not have exposure to the TI-MBRP protocol throughout the study. To ensure fidelity, interventionists self-administered a checklist of key session content after each TI-MBRP session.

Participants in both conditions were provided with audiorecorded meditation practices to use daily. A postcourse assessment was administered within 1 week of intervention completion. One week after postcourse assessments, participants were invited to focus groups to provide feedback about the intervention. Follow-up assessments were administered at 1, 3, 6, 9, and 12 months after the final intervention session. If participants were unable to be physically present, assessments were administered by phone.

Participants were given $20 gift cards for completion of each of baseline, postcourse, and focus group, and $40 gift cards for subsequent assessments. If participants completed all assessments, they received an additional $20 gift card.

Design and analyses

A cluster-randomized repeated measures design was employed to determine feasibility and acceptability of TI-MBRP. Descriptive results for feasibility and acceptability are described in text hereunder. To explore preliminary differences in symptom change, linear mixed modeling (LMM) was used to determine between-group differences in target outcomes over a 12-month follow-up period. LMM is a robust analysis due to its capability to include participants with missing data, tolerate heterogeneity at different time points, and account for variability over time within each subject.⁶⁵

Reported in text hereunder are exploratory omnibus models for condition and time × condition interactions. Parameter estimates given in Table 2 reflect differences from the mean at baseline—that is, intercept—over time, and mean differences between conditions over time. SPSS V26 was used as the statistical software for these analyses. Statistical significance was set at α = 0.05. CONSORT Checklist for cluster-randomized trials can be found in Supplementary Tables S1 and S2.

Table 2.

Fixed and Random Estimates of Primary Outcomes

Variable	Parameter	Estimate (CI)	SE	p
PTSD symptom	Fixed effects
Severity	Intercept	48.55 (43.89 to 53.22)	2.37	0.000
	Condition	−5.04 (−11.94 to 1.85)	3.49	0.151
	Time
	PC	−18.13 (−23.21 to −13.06)	2.57	0.000
	1M	−26.22 (−23.21 to −13.06)	3.10	0.000
	3M	−20.66 (−27.48 to 13.84)	3.45	0.000
	6M	−16.81 (−25.13 to −8.50)	4.22	0.000
	9M	−21.02 (−30.65 to −11.39)	4.89	0.000
	12M	−20.19 (−28.74 to −11.64)	4.33	0.000
	Time × condition
	PC	10.06 (1.92 to 18.19)	4.12	0.016
	1M	10.74 (1.53 to 19.94)	4.67	0.022
	3M	6.07 (−4.18 to 16.33)	5.20	0.245
	6M	5.98 (−5.89 to 17.85)	6.02	0.322
	9M	6.16 (−6.47 to 18.79)	6.40	0.337
	12M	5.01 (−7.01 to 17.02)	6.08	0.412
	Random effects
	Variance at intercept	79.19 (40.94 to 153.15)	26.65	0.003
	Variance at time (AR diagonal)	172.70 (133.24 to 223.85)	22.85	0.000
Craving	Fixed effects
	Intercept	21.31 (19.14 to 23.48)	1.10	0.000
	Condition	−0.81 (−4.02 to 2.40)	1.63	0.619
	Time
	PC	−11.35 (−13.89 to −8.81)	1.28	0.000
	1M	−13.48 (−16.51 to −10.46)	1.53	0.000
	3M	−13.31 (−16.67 to −9.96)	1.70	0.000
	6M	−16.26 (−20.35 to −12.18)	2.07	0.000
	9M	−11.92 (−16.65 to −7.89)	2.40	0.000
	12M	−11.95 (−16.12 to −7.78)	2.11	0.000
	Time × condition
	PC	−0.711 (−4.77 to 3.35)	2.06	0.730
	1M	1.19 (−3.36 to 5.73)	2.30	0.607
	3M	0.052 (−5.00 to 5.10)	2.56	0.984
	6M	4.78 (−1.35 to 10.91)	3.11	0.126
	9M	−0.348 (−6.54 to 5.84)	3.14	0.912
	12M	−0.616 (−6.48 to 5.25)	2.97	0.836
	Random effects
	Variance at intercept	12.71 (4.81 to 33.55)	6.29	0.044
	Variance at time (AR diagonal)	41.85 (31.74 to 55.18)	5.90	0.000

All p-values are two tailed. The fixed effects intercept represents the mean at baseline. The fixed effects condition parameter reflects the mean difference between groups at baseline. The fixed effects time parameter reflects the difference between the baseline mean and the associated parameter time point of broth groups combined. The time × condition parameter reflects mean differences between groups at the associated time point. The random effects variance and intercept parameter represent mean score variability between participants at baseline. Random effects variance at time reflects mean variability between all participants over time. AR heterogeneous covariance structure was used because data collected at further time points were assumed to be less correlated. Restricted maximum likelihood estimation was used to produce unbiased estimates of variance and covariance parameters. Random effects included subject intercepts, and time was entered as the repeated variable.

1M, 1-month follow-up; 3M, 3-month follow-up; 6M, 6-month follow-up; 9M, 9-month follow-up; 12M, 12-month follow-up; AR, autoregressive; CI, confidence interval; PC, postcourse; PTSD, post-traumatic stress disorder; SE, standard error.

Results

Feasibility

Recruitment took place between June 1, 2018, and August 30, 2019, and the trial period ended on December 31, 2020. Of 96 individuals recruited, 86% were eligible and enrolled. TI-MBRP averaged 8.5 individuals per group (range = 5–10); MBRP averaged 9 individuals per group (range = 7–10). No differences between TI-MBRP (M = 4.34; standard deviation [SD] = 2.95) and MBRP (M = 4.95; SD = 2.43) were observed in number of sessions attended. Of those randomized to TI-MRBP (n = 38), 57.90% were retained at 1-month follow-up, and 36% at 12-month follow-up, compared with MBRP (n = 45) with 62.22% at 1-month follow-up and 27% at 12-month follow-up. Approximately 33% identified as ethnic and/or racial minorities at baseline (Fig. 1 and Table 1).

FIG. 1.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the progress through the phases of a randomized-controlled trial of two groups (enrollment, intervention allocation, follow-up, and data analysis). Participants were excluded from analyses if they completed <2 assessments. 1M, 1-month follow-up; 3M, 3-month follow-up; 6M, 6-month follow-up; 9M, 9-month follow-up; 12M, 12-month follow-up; MBRP, mindfulness-based relapse prevention; PC, postcourse; TI-MBRP, trauma-integrated mindfulness-based relapse prevention.

Table 1.

Baseline Characteristic

	MBRP (n = 45)	TI-MBRP (n = 38)	Full sample (N = 83)
	M (SD)	M (SD)	M (SD)
Age	32.55 (7.86)	37.22 (10.28)	34.67 (9.27)
First trauma	8.88 (8.71)	9.68 (8.79)	9.26 (8.70)
First tobacco use	14.05 (3.75)	14.67 (6.18)	14.35 (5.02)
First alcohol use	14.02 (3.37)	14.02 (5.34)	14.02 (4.36)
First illicit substance use	16.60 (6.14)	16.36 (6.72)	16.49 (6.38)
Previous times in treatment	3.22 (1.97)	2.81 (1.82)	3.04 (1.91)
	n (%)	n (%)	N (%)
Gendera
Cisgender female	43 (95.6)	38 (97.4)	81 (98.8)
Gender neutral	1 (2.2)	0 (0)	1 (1.2)
Hispanic/Latinxa
Yes	4 (8.9)	4 (10.3)	8 (9.9)
No	40 (88.9)	33 (84.6)	73 (90.1)
Racial backgrounda
Caucasian	29 (64.4)	30 (76.9)	36 (72.8)
African American	3 (6.7)	4 (6.7)	7 (8.6)
American Indian/Alaska Native	2 (4.4)	1 (2.6)	3 (3.7)
Multiracial	5 (11.1)	1 (2.6)	6 (7.4)
Identity not listed	5 (11.1)	1 (2.6)	6 (7.4)
Sexual orientationa
Heterosexual	35 (77.8)	27 (69.2)	62 (76.5)
Lesbian	1 (2.2)	2 (5.1)	3 (3.7)
Bisexual	9 (20.0)	7 (17.9)	16 (19.8)
Relationship statusa
Single	18 (40)	7 (17.9)	25 (30.5)
Causal dating	5 (11.1)	1 (2.6)	6 (7.3)
Dating a significant person	5 (11.1)	8 (20.5)	13 (15.9)
Married/partnered	7 (20.0)	14 (35.9)	21 (25.6)
Separated/divorced	9 (15.4)	6 (15.4)	15 (18.3)
Polyamorous	0 (0.0)	2 (5.1)	2 (2.4)
Clinical setting
Residential	17 (37.8)	28 (71.8)	45 (53.6)
Intensive outpatient	28 (62.2)	11 (28.2)	39 (46.4)
Primary drug of choicea
Methamphetamines	24 (53.3)	12 (30.8)	36 (46.2)
Alcohol	9 (20.0)	15 (38.5)	24 (30.8)
Heroin	5 (11.1)	2 (5.1)	12 (9.0)
Prescription opiate medication	1 (2.2)	2 (5.1)	3 (3.8)
Cocaine	1 (2.2)	2 (5.1)	3 (3.8)
Cannabis	1 (2.2)	2 (5.1)	3 (3.8)
Crack	1 (1.1)	0 (0.0)	1 (1.2)
Other	1 (2.2)	0 (0.0)	1 (1.2)

Those randomized to MBRP were significantly more likely to endorse a “single” relationship status, “methamphetamine” as the primary substance of choice, and receive treatment in intensive outpatient settings.

^a Denotes some participants are missing from descriptive output.

MBRP, mindfulness-based relapse prevention; SD, standard deviation; TI-MBRP, trauma-integrated mindfulness based relapse prevention.

Acceptability

Acceptability was measured by course satisfaction (OCSS) and homework compliance. TI-MBRP participants rated the intervention as highly important (M = 8.14, SD = 1.85), and reported high likelihood of engaging in formal (M = 8.19, SD = 1.77) and informal (M = 8.24, SD = 1.97) mindfulness practice after the intervention. This did not significantly differ from MBRP participants (importance: M = 8.03, SD = 1.96; formal practice M = 8.05, SD = 2.04; informal practice: M = 8.14, SD = 2.31).

Regarding homework compliance, TI-MBRP participants engaged in an average of 120.90 min (SD = 126; range = 14–280) of formal practice, and 32 instances (M = 32.81, SD = 49.04) of informal practice per week at 12-month follow-up. MBRP participants engaged in an average of 127.9 min of formal practice (SD = 112.22, range = 0–371), and 34.88 instances (SD = 61.56; range = 0–169) of informal practice per week at 12-month follow-up. No significant differences were observed between conditions on formal or informal practice.

Preliminary efficacy

Participants across both conditions reported significant reductions in craving over time, F(6, 157.75) = 22.20, p = 0.000. There were no significant overall differences in craving between conditions, F(1, 72.07) = 0.021, p = 0.884, and no condition-by-time interaction, F(6, 158.75) = 0.659, p = 0.683. Effect sizes were large at 1-month follow-up in both TI-MBRP (η² = 0.912) and standard MBRP (η² = 0.881).

PTSD symptom severity significantly decreased over time, F(6, 170.39) = 14.85, p = 0.000 across both conditions, with no overall differences between conditions, F(6, 170.39) = 0.181, p = 0.672. However, parameter estimates revealed MBRP participants endorsed significantly lower PTSD symptom severity at postcourse, b = 10.06, p = 0.016, confidence interval (CI, 1.91 to 18.92), and 1-month, b = 1.29, p = 0.022, CI (0.137 to 2.46), compared with TI-MBRP. Effect sizes were large at 1-month follow-up in both TI-MBRP (η² = 0.897) and standard MBRP (η² = 0.905) (Table 2).

Discussion

Results support feasibility and acceptability of TI-MBRP in residential and IOP SUD settings. Although recruitment was short (by four participants) of this goal, the enrollment target was met; this aligns with literature indicating women in SUD treatment are interested in trauma-integrated approaches.⁴³ Using recommended strategies to retain hard-to-reach populations,^59,60 the overall 1-month retention goal of 60% was met^66,67; however, retention in TI-MBRP was below target, demonstrating higher attrition at postcourse and 1-month follow-up than in MBRP. This is consistent with high attrition rates among individuals in trauma-focused treatment.⁶⁸

Acceptability was measured through homework compliance and course satisfaction. Participants in both conditions engaged in the expected amount of weekly mindfulness practice. TI-MBRP participants reported high satisfaction, aligning with literature indicating importance of gender-focused SUD interventions.^24,69 Although MBRP was originally designed as outpatient aftercare, participants expressed appropriateness for early stage residential SUD treatment given additional supports (e.g., 24-h counselor availability). Indeed, previous research has demonstrated promise for MBIs in residential treatment settings.^49,70–72

There were overall reductions in craving across interventions over time. Since both treatment conditions addressed management of craving, outcomes were expected to differ from one another.^47,73 Furthermore, effect sizes of TI-MBRP on craving over time were similar to other integrated PTSD-SUD treatments.⁷⁴ This suggests direct discussion of trauma-related symptoms within the context of relapse prevention programs may not detract from SUD outcomes, aligning with literature advocating for PTSD-SUD treatment integration.¹³ Moreover, TI-MBRP appears to enact change in the expected direction similar to the well-validated MBRP protocol; however, these results are not confirmatory.

Both conditions showed reductions in PTSD symptom severity over the 12-month follow-up period and achieved effect sizes similar to other PTSD-SUD interventions.⁷⁴ Participants randomized to standard MBRP, versus TI-MBRP, reported greater decreases in PTSD symptom severity at postcourse and 1-month follow-up. Although both interventions resulted in sustained reductions, PTSD symptom reduction was slower among those in TI-MBRP.

Reductions in PTSD symptom severity have been noted in MBI research.^75–78 Mindfulness practices target general avoidance and negative alterations in cognitions and mood, which maintain PTSD-related intrusions and subsequent hypervigilance.⁷⁹ The relationship between PTSD symptoms and SUD has shown to be mediated by mindfulness subfacets (i.e., acting with awareness and a nonjudgment).⁸⁰ MBIs for psychological disorders have been described as a top-down approach⁸¹ in which active reorientation of awareness from automatic thoughts to the present moment is constantly being adjusted.

Perhaps repeated reorienting from habitual patterns of thinking to sustained awareness of physiological sensations extinguishes ongoing trauma-related rumination, in turn reducing associated emotional distress and physiological arousal^35,82 without direct discussion of trauma-related content. The results provide preliminary evidence that standard MBRP, without direct discussion of trauma, may be sufficient to address both PTSD and SUD symptoms. These exploratory outcomes should be confirmed in a follow-up study with a larger sample.

Limitations

This study contains several limitations. Although LMM is recommended for repeated measures designs with small sample sizes and high attrition,⁸³ results reflecting symptom change should be interpreted as preliminary, as the study was powered to assess feasibility outcomes at 1-month follow-up, and not to detect within- and between-group differences at longitudinal time points. Second, assessments were self-report and susceptible to social desirability bias.^84,85 Sensitive information about substance use was collected while participants were still in treatment, which may have resulted in under-reporting to maintain treatment privileges.

Future trials should administer urine tests to corroborate self-reported substance use. Severity of dependence before treatment should also be collected to control for varying use and dependence. Polysubstance use should also be taken into account, as PTSD symptoms and craving can function differently for various drugs of choice.^86,87 Furthermore, future trials might track mindfulness practice through smartphone application for more objective practice measurement.⁸⁸

Conclusions

Results from this study suggest that TI-MBRP is safe and acceptable for women with PTSD-SUD in both residential and IOP settings. Although feasibility was supported by these findings, future efforts should be taken to increase retention at distal follow-up points. Reduction of PTSD symptoms and craving was demonstrated in both groups. TI-MBRP was not superior to standard MBRP in PTSD symptom severity outcomes at any time point. This indicates adaptations to MBRP may not be necessary to address both PTSD and SUD outcomes, as standard MBRP may already address mechanisms underlying both disorders.

However, further refinement of TI-MBRP may be warranted as trauma- and gender-focused treatments are rarely implemented in community-based SUD programs.⁸⁹ As women often endorse preference for integrated PTSD-SUD treatment,⁴³ trauma- and gender-focused MBIs may offer treatment-seeking women an evidence-based PTSD-SUD approach. Moreover, future studies with sufficient power to detect within- and between-group differences may assess efficacy of MBRP in comparison with other EBIs for PTSD-SUD to determine its suitability as an alternative for individuals who prefer nontrauma-focused treatments.

Authors' Contributions

Both authors made significant contributions to this study and the article. V.C.S. was the principle investigator of this study and was involved in intervention adaptation, recruitment and consenting participants, training group facilitators and delegating study tasks to research assistants, data collection and preparation, and writing the article. S.B. was the faculty research mentor for this project and oversaw all study procedures. She also made significant contributions and edits to the final article.

Author Disclosure Statement

All authors on this article have no competing interests, financial interests, or conflicts of interest in funding or employment on account of this submission. The views expressed in the submitted article are of the mentioned authors and not an official position of their respective institutions.

Funding Information

This study was supported by the National Institute on Drug Abuse R36 grant (No. 1R36DA043554-01) and the Mind and Life Institute Varela grant (No. A-38168754). No other sources of support that facilitated conduct of the study described in the article or the writing of the article itself were used.