Dispositional mindfulness and trauma symptom severity associated with uptake of mindfulness practice among women in treatment for substance use disorder

ABSTRACT

Objective

This study assessed the preintervention predictors of participant initial satisfaction and uptake of mindfulness practices during an MBI (Moment-by-Moment in Women’s Recovery).

Methods

Data from 100 women (M_age = 32.38 years, 60% Latina) in residential SUD treatment randomly assigned to the MBI were used to assess baseline participant demographics and clinical characteristics and three acceptability variables (satisfaction and formal and informal mindfulness practice) after the second and third of 12 intervention sessions, respectively. Results: Multiple linear regression analyses found that the frequency of formal mindfulness practice was associated with preintervention mindfulness disposition (β = 0.245, p < .05) and age (β = .266, p < .05; R² = .137, F[4, 77] = 3.047, p < .05). The frequency of informal mindfulness practice was associated with preintervention mindfulness disposition (β = .239, p < .05), and preintervention trauma symptom severity was significant (β = .226, p < .05; R² = .139, F[4, 77] = 3.116, p < .05).

Conclusions

It is important to consider participant, and intervention characteristics may be associated with participant adoption of intervention teachings and practices early in intervention implementation. Identifying these characteristics and adapting the intervention accordingly is necessary for the long-term goal of improving intervention acceptability among vulnerable populations at high risk of treatment dropout.

KEYWORDS:

• Mindfulness practice
• satisfaction
• substance use disorders
• residential
• recovery
• women
• trauma

Introduction

Findings from research on mindfulness-based interventions (MBI) for individuals in treatment for substance use disorders (SUD) are promising despite still being in the early stages (Garland & Howard, 2018; Li et al., 2017). Yet an important and understudied factor in MBIs is the acceptability and potential fit of MBIs for a broad spectrum of individuals in SUD treatment. Acceptability is a multifaceted construct that reflects the degree to which participants find an intervention to be appropriate (Sekhon et al., 2017). Acceptability is regarded as important for intervention retention, adoption, implementation, and dissemination (Bak et al., 2018; Diepeveen et al., 2013; Proctor et al., 2011; Stok et al., 2015).

To date, research on acceptability specifically related to MBIs for SUD has been extremely limited and demonstrates an inadequate assessment of variations in acceptability across participant profiles (Bautista et al., 2019). A recent review (Bautista et al., 2019) of MBIs for SUD found that only four of 17 studies assessed acceptability (Amaro et al., 2014; Bowen et al., 2009, 2017; Witkiewitz et al., 2013). The instruments used to assess acceptability were session attendance (Amaro et al., 2014), participant satisfaction (Amaro et al., 2014; Bowen et al., 2017), uptake of mindfulness practice (Bowen et al., 2009), and 15-week follow-up (Witkiewitz et al., 2013). Questions remain unanswered regarding the construct validity of assessing acceptability based on completion rate or session attendance (Bautista et al., 2019). It is possible for participants to find an intervention acceptable without being able to attend all sessions or complete the intervention due to reasons unrelated to acceptability (e.g., scheduling conflicts, dropping out of treatment, transferring to another treatment program, illness). Therefore, satisfaction and mindfulness practice measures are potentially more accurate assessments of acceptability. The most common measures of mindfulness practice include assessments of formal and informal mindfulness practice. Formal practice refers to designated time engaging in a specific type of mindfulness activities, such as a body scan, walking meditation, or sitting meditation. Informal practice refers to intentional mindfulness practice and awareness of thoughts, feelings, and bodily sensations throughout ordinary activities, daily life, or challenging situations (Birtwell et al., 2019; Enkema & Bowen, 2017).

Factors that may be associated with acceptability

The mindfulness-based relapse prevention (MBRP) intervention was designed as an aftercare SUD program because the program is best suited for individuals who completed inpatient or outpatient treatment, have clarity in their goals, and understand how to comply with the treatment program (Bowen et al., 2011). MBRP was adapted from mindfulness-based stress reduction (MBSR), which is an evidence-based mindfulness training program aimed at reducing stress, anxiety, depression, and pain (Kabat-Zinn, 1990). In a qualitative study of MBRP participants in a treatment program that offered residential and day programs for SUD, Harris (2015) referenced participant responses to the intervention based on the recovery phase, noting that for participants in the early stages of recovery, MBRP sessions were challenging, and they found it difficult to confront their emotions and cravings. To address the potential concerns of service providers in introducing an MBI during treatment, a new intervention was developed with multiple curriculum adaptations and continued input from SUD clinicians and clients. The result was Moment-by-Moment in Women’s Recovery (MMWR), which focused on the connection between stress and relapse and was suitable for women who were early in their recovery, ethnoculturally diverse, and had low literacy. Initial assessment of MMWR showed high participant satisfaction (Vallejo & Amaro, 2009). However, the question of whether time in SUD treatment is associated with participant acceptability remains largely unexplored.

A second possible predictor of acceptability of MBIs is mindfulness predisposition, a traitlike propensity that “reflects a greater tendency to abide in mindful states over time” (Brown et al., 2007, p. 218). Mindfulness predisposition has been positively associated with greater self-control (Lakey et al., 2007) and greater autonomous motivation for daily tasks and behavior changes (Brown & Ryan, 2003). Thus, mindfulness predisposition may also be associated with participants’ initial response to MBIs and uptake of mindfulness practice (behavior change) because they may find it easier to practice mindfulness and perceive greater benefits.

Posttraumatic stress disorder (PTSD) symptom severity, which is highly prevalent among women in SUD treatment (Du et al., 2013), may be associated with initial acceptability of MBIs for SUD (M. M. Kelly et al., 2012). Safety concerns (e.g., triggering traumatic memories) about the use of non-trauma informed and adapted MBIs with individuals with co-occurring mental health disorders, particularly the history of trauma, were raised early in the literature (Dobkin et al., 2012; Kelly, 2015; Vallejo & Amaro, 2009). However, more recent evidence suggests overlapping mechanisms of action, through which mindfulness may be effective in aiding the recovery process for both SUD and trauma (María-Ríos & Morrow, 2020; Vujanovic et al., 2020). Further, MBIs are specifically designed for individuals with trauma, and evidence suggests that MBIs are not only safe for individuals with PTSD symptoms but also that mindfulness practices may be helpful in treating PTSD symptoms (Kelly, 2015; Kelly & Garland, 2016). Skills such as the capacity to regulate attention and traumatic thoughts could improve control over arousal and, therefore, enable a new way to process traumatic memories with metacognitive awareness (Kelly & Garland, 2016). However, little is known about the uptake of formal and informal mindfulness practice and satisfaction with MBIs for SUD among individuals with PTSD symptoms, or if symptom severity is associated with greater uptake of mindfulness practices.

To address gaps in the research on acceptability, this study aimed to assess the predictors of three acceptability variables (measured using surveys of satisfaction and frequency of formal and informal mindfulness practice) of a 12-session MBI (MMWR) designed for ethnoculturally diverse women in SUD treatment (Vallejo & Amaro, 2009). The MMWR program focuses on the role of stress in relapse while integrating issues of trauma and mental health problems and their effects on relapse (Vallejo & Amaro, 2009). In the framework of trauma-informed care (Amaro et al., 2005; Bowen & Murshid, 2016), part of the adaptation also involves modifying aspects of some mindfulness practices to increase a sense of safety for women with trauma-related symptoms and mental health difficulties (Muskett, 2014; Raja et al., 2015; Rosenberg, 2011). We hypothesized that (1) time in treatment prior to the start of the intervention would be positively associated with satisfaction, frequency of formal mindfulness practice, and frequency of informal mindfulness practice; (2) greater mindfulness predisposition at baseline would be positively associated with satisfaction, frequency of formal mindfulness practice, and frequency of informal mindfulness practice; and (3) greater PTSD symptom severity at baseline would be positively associated with satisfaction, frequency of formal mindfulness practice, and frequency of informal mindfulness practice.

Method

Participants

All participants were adult women admitted to the residential SUD treatment program study site in Southern California and clinically diagnosed with SUD based on the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5; American Psychiatric Association, 2013). The present study included only participants randomized to the MMWR intervention condition (see, Amaro & Black, 2017 for randomization procedure). Inclusion criteria were client at the residential treatment study site, female, aged 18–65 years, diagnosed with SUD, fluent in English, and agreed to participate in the study. Exclusion criteria were inability to understand or sign the informed consent due to language reasons, cognitive impairment, untreated psychotic disorder, untreated chronic severe mental health condition, past-30-day suicidality based on clinical intake assessment, more than 65 years old, current prisoner, more than 6 months pregnant, enrolled in another study, not willing to sign a HIPAA form or be audio recorded during interviews and intervention sessions, and missing the first intervention orientation session. Table 1 provides participant characteristic information. For more information about the sample characteristics, see Black and Amaro (2019).

Table 1. Participants’ descriptive information (N = 100).

Variable	M (SD)	n (%)
Age	32.38 (9.82)
Race and ethnicity
Hispanic		60 (60.00)
Non-Hispanic Black		18 (18.00)
Non-Hispanic White		20 (20.00)
Other		2 (2.00)
Education (years)	11.67 (2.15)
Days from treatment entry to MMWR start	37.35 (15.86)
Mindfulness predisposition^a	76.96 (12.74)
Trauma severity^b	16.23 (11.94)
Number of sessions attended (of 12)	9.49 (3.20)
Satisfaction	3.98 (0.72)
Formal practice	1.97 (1.06)
Informal practice	2.34 (1.24)

^aMindfulness predisposition = total score on the Five Factor Mindfulness Questionnaire; possible scores range from 25 to 125; and higher scores indicate more mindfulness predisposition.

^bTrauma severity = total score on the PTSD Symptom Scale–Self Report; possible scores range from 0 to 68; and higher scores indicate higher trauma severity.

Procedure

An on-site study coordinator identified female patients who met study eligibility criteria using information from the residential treatment site’s intake assessment. The study coordinator then confirmed eligibility and obtained permission from eligible patients to be contacted by the study interviewer, who made appointments with prospective participants, conducted the informed consent, and administered the baseline assessment (Amaro & Black, 2017). Participation in the study was voluntary. Additional information regarding procedures can be found in an article published by the principal investigators of the parent study (Amaro & Black, 2017).

Moment-by-Moment in Women’s Recovery

MMWR was delivered twice weekly for 80 minutes each session for 12-group sessions over 6 weeks. The intervention took place during participants’ residential treatment. The MMWR teachers were trained in both MBSR and MMWR and facilitated the session with a trained on-site master’s-level clinician with experience in SUDs. The teachers were guided by an instructional manual with standardized lesson plans (Amaro & Black, 2017; Black & Amaro, 2019). Information about teacher training and fidelity of intervention delivery is presented elsewhere (Amaro & Black, 2017; Black & Amaro, 2019).

Measures

Time in treatment prior to intervention

Obtained from clinical records, this score was calculated as the number of days between treatment entry and the first MMWR session, with a range of 5–74 days (M = 37.35, SD = 15.86); 78% of the samples had been in treatment for at least 3 weeks prior to starting MMWR.

Mindfulness predisposition

The Five-Factor Mindfulness Questionnaire has 25 items (response categories ranging from 1 = never or rarely to 5 = always) and was completed at baseline assessment to assess mindfulness disposition prior to intervention start. A sample item is “When I had distressing thoughts or images (that bothered me), I just noticed them without reacting.” The measure has demonstrated validity (Baer et al., 2006) and internal consistency with individuals’ transitioning out of intensive SUD treatment (α = .91; Bowen et al., 2009). For the current sample, Cronbach’s alpha for the total scale was α = .83.

PTSD symptom severity

The PTSD Symptom Scale–Self Report, composed of 17 items (rated from 0 = not at all to 3 = almost always) was administered at baseline. Reliability and validity have been shown for assessing PTSD symptoms experienced by the participants in the last month (Foa et al., 2005). A sample item is “How often have you been bothered by having bad dreams or nightmares about the traumatic events?” This scale measures the frequency of reexperiencing, avoidance, and arousal symptoms related to trauma exposure over the past 30 days. For the current sample, the total scale score was used, and Cronbach’s alpha was α = .93.

Satisfaction survey

Satisfaction data were collected at the end of session 2 of 12 sessions. The satisfaction survey consisted of 17-items rated from 1 (not at all) to 5 = (very much), with high scores indicating higher satisfaction. Items assessed various aspects of satisfaction: session content, skills learned, perceived usefulness, and importance for recovery. This survey was developed based on a previous feasibility study and designed to match the MMWR curriculum. For the current sample, Cronbach’s alpha was α = .95.

Frequency of formal mindfulness practice survey

Formal mindfulness practice frequency data were collected by study research staff (without the facilitator in the room) at the end of session 3 of 12 sessions. The formal mindfulness practice survey consisted of six items rated from 0 (never) to 5 (4 or more times a day) that assessed the frequency of use of specific types of formal mindfulness practices (e.g., sitting and walking meditation, love and kindness mediation, and mindful stretching) since the previous class session. This survey was developed based on a previous feasibility study and designed to match the MMWR curriculum; the full survey is included in Table 2. For the current sample, Cronbach’s alpha was α = .80.

Table 2. Mindfulness practice survey.

In the last seven days …	Never	Less than once a day	Once a day	2 times a day	3 times a day	4 or more times a day
Formal practice
How often did you practice or use sitting meditation with the audio?	○	○	○	○	○	○
How often did you practice or use sitting meditation without the audio?	○	○	○	○	○	○
How often did you practice Love and Kindness meditation?	○	○	○	○	○	○
How often did you practice walking meditation?	○	○	○	○	○	○
How often did you practice or use the body scan?	○	○	○	○	○	○
How often did you practice or use mindful stretching?	○	○	○	○	○	○
Informal practice
How often did you practice or use the Stop Light Technique?	○	○	○	○	○	○
How often did you practice or use mindfulness in noticing your breath?	○	○	○	○	○	○
How often did you practice or use mindfulness to be aware of your emotions?	○	○	○	○	○	○
How often did you practice or use mindfulness to be aware of your thoughts?	○	○	○	○	○	○
How often did you practice or use mindfulness to be aware of your body sensations like your heartbeat, sweaty hands, pain, other?	○	○	○	○	○	○
When I have cravings, I used mindfulness to notice my cravings without judgment	○	○	○	○	○	○
When I have craving, I used mindfulness to experience cravings without reacting	○	○	○	○	○	○
How often did you practice or use mindfulness or anything else you learned in class for something else in your life?	○	○	○	○	○	○

Informal mindfulness practice survey

Informal mindfulness practice frequency data were collected by study research staff (without the facilitator in the room) at the end of session 3 of 12 sessions. The informal mindfulness practice survey consisted of eight items rated from 0 (never) to 5 (4 or more times a day) that assessed the frequency of use of specific types of informal mindfulness practices (e.g., awareness of emotions, thoughts, body sensations, and cravings) since the previous class session. This survey was developed based on a previous feasibility study and designed to match the MMWR curriculum; the full survey is included in Table 2. For the current sample, Cronbach’s alpha was α = .92.

Data analyses

Bivariate Pearson correlations were calculated to test associations between variables of interest and whether age and education needed to be included as covariates. Multiple linear regression models were conducted to examine days of treatment prior to the start of MMWR, baseline mindfulness predisposition, and baseline PTSD symptom severity as independent variables in three models using early intervention satisfaction, early uptake of formal mindfulness practice, and early uptake of informal mindfulness practice as dependent variables, respectively.

Results

A significant positive correlation existed between age and formal mindfulness practice frequency (r = .252, n = 82, p < .05). Therefore, age was included as a covariate in the regression model with formal mindfulness practice frequency as the dependent variable. There is a marginally significant positive correlation between education and informal mindfulness practice frequency (r = .216, n = 82, p < .10). Therefore, education was included as a covariate in the regression model with informal mindfulness practice frequency as the dependent variable.

Regression

To test the hypothesized associations between variables of interest, three regression models were conducted (see, Table 3). The first regression model included mindfulness predisposition, days of treatment prior to the start of the intervention, and PTSD symptom severity as independent variables. This model explained 5% of the variance in satisfaction at session 2 and was not significant (R² = .048, F [3, 86] = 1.458, p  =.20). None of the three independent variables provided a significant unique contribution to participants’ satisfaction with the intervention. The second model, which included age, days in treatment prior to the start of the intervention, mindfulness predisposition, and PTSD symptom severity, explained 14% of the variance in formal mindfulness practice frequency at session 3 (R² = .137, F [4, 77] = 3.047, p < .05). Results indicate that older age (β = .266, p < .05) and greater PTSD symptom severity (β = .245, p < .05) were both positively and significantly associated with greater formal mindfulness practice frequency. The third model included education as a covariate and days of treatment prior to the start of the intervention, mindfulness predisposition, and PTSD symptom severity as independent variables. This model explained 14% of the variance in informal mindfulness practice frequency at session 3 (R² = .139, F [4, 77] = 3.116, p < .05). Results indicate that greater mindfulness predisposition (β = .239, p < .05) was associated with more informal mindfulness practice frequency, whereas PTSD symptom severity (β = .226, p < .05) was a marginally significant predictor in the same model.

Table 3. Multiple regression analyses of intervention and participant characteristics as predictors of acceptability variables.

	Satisfaction					Formal Practice					Informal Practice
	b	SE	β	p	CI	b	SE	β	p	CI	b	SE	β	p	CI
Age						0.029	0.012	.266	.015	.006, .052
Education											0.091	0.060	.164	.132	−.028, .211
Days in treatment	0.007	0.005	.151	.172	−.003, .017	0.007	0.007	.110	.331	−.007, .022	0.011	0.009	.138	.224	−.007, .028
Mindfulness predisposition	−0.001	0.006	−.026	.817	−.014, .011	0.014	0.009	.178	.123	−.004, .033	0.023	0.011	.239	.041	.001, .047
Trauma severity	−0.008	0.007	−.138	.222	−.022, .005	0.022	0.010	.245	.034	.002, .042	0.024	0.012	.226	.050	.001, .047
	R^2 = .048, F(3, 86) = 1.458, p  .20					R^2 = .137, F (4, 77) = 3.050, p < .05					R^2 = .139, F(4, 77) = 3.120, p < .05

Discussion

This study investigated the relationships between participant characteristics and three variables of early intervention acceptability: satisfaction, frequency of formal mindfulness practices, and frequency of informal mindfulness practices. Informal mindfulness practice frequency was positively associated with baseline mindfulness predisposition and had a marginally significant association with PTSD symptom severity, whereas formal mindfulness practice frequency was only positively associated with PTSD symptom severity. Days in treatment prior to the intervention, PTSD symptom severity, and mindfulness predisposition were not significantly associated with participant satisfaction with the intervention.

There was a significant positive relationship between mindfulness predisposition and the frequency of informal mindfulness practices. This indicates that individuals with greater mindfulness predisposition prior to starting the intervention had greater uptake of informal mindfulness practices by the end of the third session. Although previous research did not assess the relationship between baseline mindfulness predisposition and mindfulness practice during the intervention, Kiken et al. (2015) found that state mindfulness was higher immediately following a meditation session and that repeating these sessions for 7 weeks was associated with higher mindfulness disposition, suggesting a synergistic relationship between mindfulness disposition and mindfulness practice. While mindfulness predisposition is sometimes thought of as a trait, it is malleable with greater exposure to mindfulness skills training (Kiken et al., 2015). The results of the present study may suggest that uptake of informal mindfulness practice, as a malleable skill, can be enhanced by providing participants who score low on mindfulness disposition with a preintervention readiness module or additional sessions to work on developing mindfulness practice over time through a scaffolding process to divide the learning into smaller incremental chunks of information to enhance learning and adoption of mindfulness practices via small steps that build on the tools provided throughout the behavior change process.

PTSD symptom severity was significantly associated with frequency of uptake of formal mindfulness practices and had a marginally significant association with informal mindfulness practices, but it was not a significant predictor of satisfaction. This may suggest that women with higher PTSD symptom severity were using mindfulness practices more frequently outside of class to cope with PTSD symptoms, substance cravings, or other stressors related to SUD. Participants experiencing more severe PTSD symptoms may be more motivated to adhere to treatment due to a recognition that they may need it more or benefit more than patients with less severe PTSD symptoms. Vujanovic et al. (2020) found that higher baseline mindfulness was associated with lower trauma severity following 12-week cognitive-behavioral therapy for SUD. The authors explained this result by suggesting that individuals with greater mindfulness may be more aware of emotional and physiological states associated with trauma symptomatology, which may make them better able to accept negative emotions and focus on cognitive processing during treatment (Vujanovic et al., 2020). Findings such as these indicate the need for further investigation of the possible relationships between mindfulness predisposition, mindfulness practice (frequency and quality), and trauma (severity and symptomology). This finding may be specific to MMWR and is likely due, at least in part, to the intervention design adapted to address the role of trauma in the recovery process.

Time spent in treatment prior to starting MMWR ranged from 5 to 74 days, with an average of slightly more than a month (37.35 days). Only 21% of the sample had been in treatment for 3 weeks or less. Therefore, most of the sample was likely already adjusted to the treatment facility when starting the intervention. Future studies may consider using a variable that captures participants’ readiness for the intervention.

Limitations and future directions

Although the acceptability variables of satisfaction and frequency of formal and informal mindfulness practices demonstrated adequate content validity and strong internal consistency, more rigorous psychometric testing could be performed to conclude the validity and reliability of the measures. There is currently no gold-standard tool to assess acceptability of MBIs for SUDs (Bautista et al., 2019). The acceptability measures used in the present study were developed through multiple iterations with previous intervention participants. As with all studies that evaluate intervention satisfaction and uptake of behavioral practices, it is important to acknowledge the potential effect of social desirability on reports of satisfaction and frequency of formal and informal mindfulness practices. To reduce this potential influence, the facilitators were not in the room.

Future studies could investigate possible predictors of satisfaction; the present study found that mindfulness disposition, PTSD symptom severity, and time in treatment prior to starting the intervention were not associated with satisfaction. Prior research found that being mandated to treatment also did not predict satisfaction (J. F. Kelly et al., 2005). Factors such as optimism toward recovery and intervention efficacy, craving of substances, depression or other mental distress, and spirituality have previously been assessed as outcome measures in MBIs for SUD (Li et al., 2017). However, these variables assessed at baseline may influence satisfaction with the intervention. Satisfaction may also be a mediating variable to help explain the pre–post change score among some constructs. Future studies may also consider assessing the relationship between acceptability variables and intervention outcomes immediately following the intervention and in the long term.

Overall, this research adds to the existing literature by providing an in-depth assessment of acceptability of an MBI among women with SUD and PTSD symptom severity. Our findings suggest that mindfulness practice frequency was positively associated with participants’ baseline PTSD symptom severity and mindfulness predisposition. Prior research on acceptability has been cursory, and there is very limited information regarding how acceptability is associated with other intervention and implementation factors. Identifying these factors and adapting the intervention accordingly is necessary for the long-term goal of improving intervention acceptability among vulnerable populations at high risk of treatment dropout.

Authors’ contribution

Both authors equally contributed to writing the manuscript. The first author ran the data analyses. The second author is a principal investigator for the parent study.

Compliance with ethical standards

The University of Southern California Institutional Review Board approved this study (UP-14-00391). All persons gave informed consent prior to their inclusion in the study.

Registered in clinicaltrials.gov (NCT02977988)

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

NIDA grant # [5R01DA038648]; NIDA grant # [5R25DA026401]; NCATS grants # [5TL1TR001864].