Perspectives on trauma and the design of a technology-based trauma-informed intervention for women receiving medications for addiction treatment in community-based settings

Abstract

Background:

Despite national calls to develop gender-specific interventions for women with opioid use disorder (OUD) with co-occurring trauma and post-traumatic stress disorder (PTSD) symptoms, there remains a dearth of research on what modalities or treatment components would be most feasible for this population. This study interviewed women with OUD receiving medication assisted treatment and addiction treatment providers to explore (a) experiences of barriers to receiving trauma treatment, and (b) both the perceptions and desired design of a prospective technology-delivered, trauma-informed treatment for women with OUD.

Methods:

Women with lifetime OUD (n = 11) and providers (n = 5) at two community substance use clinics completed semi-structured interviews. Interviews were transcribed, coded, and analyzed in NVivo v11 using a grounded theory approach. Women also completed a demographic form and clinical measures.

Results:

Clients were primarily women with children reporting histories of multiple trauma exposures, high PTSD symptoms, and polysubstance use. Two themes emerged among clients and one among providers regarding barriers to trauma treatment. Regarding the feasibility and desired attributes of a technology-based intervention, six themes emerged among clients and providers, respectively.

Conclusions:

Themes suggest a high interest by clients and providers for a technology-delivered, trauma informed treatment available by smartphone. Utilizing technology as an adjunct to care, without reducing face-to-face therapy, was important to both clients and providers.

1. Introduction

Opioid use disorders (OUD) among women in the United States are on the rise (Meyer, Isaacs, El-Shahawy, Burlew, & Wechsberg, 2019). From 1999 to 2010, heroin use among women increased by 100% in comparison to 50% among men (Center for Disease Control [CDC], 2015), and from 1999 to 2016 the number of women dying from prescription opioid overdoses increased at a faster rate than men (National Institute on Drug Abuse [NIDA], 2018). Part of the surge in opioid use among women is the result of more frequent health-seeking behaviors, higher rates of chronic pain, and social norms that make it acceptable for women to request more help when experiencing pain compared to men (Cicero, Lynskey, Todorov, Inciardi, & Surratt, 2008). Greater access to opioids has thus contributed to greater risk of misuse. Among 29,906 women and men, Green, Grimes Serrano, Licari, Budman, and Butler (2009) found that women reported significantly greater misuse of prescription opioids in the past month compared to their male counterparts. Moreover, in the state of New Jersey—the location of this study—opioid overdose deaths increased >576% among women from 1999 to 2017 (75 deaths in 1999 to 507 deaths in 2017) in comparison to 490% among men (248 deaths in 1999 to 1462 deaths in 2017; The Henry J. Kaiser Family Foundation, 2019). Substantial research has demonstrated that due to critical biological, behavioral, and epidemiological differences in the development and characteristics of substance use disorders among men and women, gender specific treatment is both recommended and beneficial to treatment outcomes (Greenfield, Back, Lawson, & Brady, 2010; Substance Abuse and Mental Health Services Administration [SAMHSA], 2016). Women with substance use disorders are more likely to have co-occurring mental health diagnoses, faster progression to use (i.e., telescoping), show biological differences in stress reactivity and gonadal hormones that impact addiction, and enroll at a lower rate in treatment all of which, further suggest a need for gender-specific interventions that may enhance treatment engagement (Greenfield et al., 2010). Accordingly, there is an imminent need for gender-specific interventions for women with OUD based on studies identifying the gender differences in opioid use and OUD treatment (Marsh, Park, Lin, & Bersamira, 2018; Meyer et al., 2019; Serdarevic, Striley, & Cottler, 2017).

Exposure to a traumatic event and the experience of its related sequelae, such as post-traumatic stress symptoms and post-traumatic stress disorder (PTSD), are highly correlated with substance use and OUD specifically. As Ecker and Hundt (2018) noted in their review, one twin study showed that when one twin was exposed to child sexual abuse there was a 2.6 increased likelihood to develop an OUD, the second highest likelihood ratio among all substances (Nelson et al., 2006). In another study, the co-occurrence of PTSD with OUD showed the highest prevalence rate of 33.2%, a value higher than all other substances when co-occurring with PTSD (Mills, Hlth, Teesson, Ross, & Peters, 2006). To complicate this picture further, women show an increased risk of trauma exposure and traumatic stress symptom development in comparison to men. Women are more likely to be exposed to severe forms of interpersonal trauma (e.g., interpersonal violence) and show sex differences in biological systems (e.g., gonadal hormones, stress response systems, and neuroendocrine profiles) that increase their vulnerability to develop PTSD (López-Castro, Saraiya, & Hien, 2017; Olff, Langeland, Draijer, & Gersons, 2007). Consequently, it is no surprise that one scoping review found women misusing prescription opioids were more likely to show a trauma history, use opioids to cope with trauma, and often began misusing prescription opioids due to negative health sequelae from traumatic events, such as chronic pain and physical health conditions (Hemsing, Greaves, Poole, & Schmidt, 2016). Failure to address trauma and PTSD in OUD treatment has dire effects. One study found that with every 10% increase in PTSD symptom severity there was an associated 36% reduction in methadone program attendance (Peirce, Brooner, King, & Kidorf, 2016). This points to an urgent need to tailor interventions to women with trauma histories and OUD presentations, as indicated at a national level (Office on Women's Health, 2017) and among many empirical articles (Ecker & Hundt, 2018; Hemsing et al., 2016; Marsh et al., 2018; Meyer et al., 2019; Sugarman, Meyer, Reilly, & Greenfield, 2019; Terplan, Hand, Hutchinson, Salisbury-Afshar, & Heil, 2015).

Technology-based interventions are a promising, feasible, and cost-effective method to supplement traditional community-based substance use treatment for OUD. They have the potential to bolster concrete self-care and recovery skills to manage symptoms of trauma and substance use (Schmitt & Yarosh, 2018). By allowing for personalized, confidential, 24/7 access to evidence-based behavioral interventions, especially for lower income communities, technology-based interventions address common barriers to care (Bickel & Marsch, 2007; Carroll & Kiluk, 2017; Dempsey, Dowling, Larkin, & Murphy, 2016; Marsch, 2012; Schmitt & Yarosh, 2018). A recent review found seven studies reporting the results of randomized controlled trials for tech-based interventions for PTSD and SUD and four papers addressing the feasibility or protocol of such interventions (Gilmore, Wilson, Skopp, Osenbach, & Reger, 2017). However, the majority of these studies focused on veterans, a predominantly male population, and on treatment for one disorder rather than both concurrently. Further, none of these studies focused specifically on OUD (Gilmore et al., 2017) even though in a different study, 88% (N = 178) of individuals receiving medications for OUD reported using the internet for health-related information (Masson, Chen, Levine, Shopshire, & Sorensen, 2019), and clients using opioids have successfully used tech-based interventions as psychosocial adjuncts for chronic pain (Wilson, Roll, Corbett, & Barbosa-Leiker, 2015). Moreover, in a qualitative investigation of barriers to substance use treatment, providers expressed lack of integrated mental health and SUD services in addition to the absence of technological adjuncts as two major hindrances to successful recovery (Ashford, Brown, & Curtis, 2018), which is especially notable given that clients expressed a desire for technology-based adjuncts (Schmitt & Yarosh, 2018). Indeed, only one study to our knowledge has piloted a psychoeducational tech-based intervention for women with SUD and found high feasibility in addition to client interest for more trauma-informed modules (Sugarman et al.,2019). Despite this growing interest and need, there are no published technology-based applications for women with trauma histories and OUDs.

Consequently, there is a pressing need to develop trauma-informed, gender-specific interventions for women with OUD and PTSD (Hemsing et al., 2016). This study gathered qualitative feedback from women receiving treatment for OUD in an outpatient setting and from addiction treatment providers to explore their experience with trauma-informed care and barriers to trauma treatment, as well as attitudes toward using a trauma-informed, technology-based intervention as an adjunctive treatment to medications for addiction treatment (MAT). Women clients and providers were viewed as the best key informants to understand experiences with trauma treatment and the potential utility of a trauma-informed technology-based intervention. A major focus of this study, in alignment with participatory designs, was to interview key expert stakeholders to explore their perceptions of trauma treatment and a potential technology intervention to empower and increase their voice in the design of such a software (e.g., Schmitt & Yarosh, 2018).

2. Methods

2.1. Participants

A total of 11 clients and 5 providers (2 psychiatrists, 1 therapist, 1 social worker, and 1 nurse) were interviewed at two community substance use treatment facilities in suburban New Jersey. Clients and providers were recruited through the distribution of study flyers, which included contact information for the project coordinator. The project coordinator went onsite to discuss the study with providers to enhance recruitment of both clients and providers. Providers advertised the study to clients meeting inclusion criteria. Interested clients and providers contacted the project coordinator for a brief phone screening to assess eligibility.

Clients were included in the study if at the time of the telephone screening they reported: (a) a current or lifetime diagnosis of OUD (confirmed through verification with providers), (b) self-reported adherence to MAT (e.g. buprenorphine, naltrexone, or methadone medication), (c) trauma exposure and symptoms of PTSD or another anxiety disorder, (d) were between the ages of 18–64 years old, (e) spoke fluent English, and (f) were currently receiving services from one of two community substance use treatment centers. Provider inclusion criteria consisted of: (a) work with women with OUD and trauma, anxiety, and/or PTSD; (b) working at one of two recruitment sites; and (c) above the age of 18. If inclusion criteria were met, an interview date was scheduled. It is important to note that clients were formally diagnosed by providers in the community substance use treatment facility for OUDs and then referred to the study. During the interview, the interviewer completed additional quantitative assessments with clients to ascertain which diagnostic criteria for an OUD they met and their symptoms of PTSD, anxiety, and/or depression (see below, Measures). All study procedures were approved by the Rutgers University Institutional Review Board, and clients and providers provided written informed consent prior to their respective interviews.

2.2. Assessment

Clients and providers met once with a trained interviewer. For clients, interviews consisted of a 30-minute quantitative assessment administered verbally by the project coordinator (see Measures) followed by a semi-structured interview that ranged from 33 min to 92 min (M = 59.74, SD = 17.63). Quantitative measures assessed for demographics, substance use history and criteria, trauma history, use of technology, and symptoms of depression, anxiety, and PTSD. For providers, semi-structured interviews ranged from 23 min to 83 min (M = 56.46, SD = 22.78).

Semi-structured interview guides were developed through discussion with the protocol team (all co-authors). Two interviewers completed a training on skills in qualitative interviewing. Open-ended questions and relevant probes were used to elicit participant feedback. Primary domains were the same for clients and providers and included: access and utilization of trauma-informed behavioral treatment as an adjunct to MAT; strengths and weaknesses of currently available behavioral treatment; current use of any technology-based interventions; barriers to the use of technology-based interventions; and familiarity with trauma-informed behavioral interventions. Upon completion of study procedures, clients were paid $50 and providers were paid $100 through gift cards.

2.3. Measures

A series of assessments were used to capture demographic and clinical variables.

2.3.1. Demographics

Authors created a demographic questionnaire assessing for age, gender, race/ethnicity, housing, criminal involvement, education, and childcare responsibilities.

2.3.2. Trauma exposure and PTSD

Lifetime exposure to traumatic events was measured by the 17-item, Life Events Checklist-5 (LEC-5; Weathers et al., 2013). Summed scores were tallied for any events which participants endorsed “it happened to me.” In addition, the total frequencies and percentages of accidental, intentional, and other lifetime traumas across the female sample were provided. The LEC has shown strong convergent validity with measures assessing for post-traumatic stress symptoms on a sample of combat veterans seeking a PTSD evaluation, (e.g., the Clinician Administered PTSD Scale, the Mississippi Scale for Combat-Related PTSD, and the PTSD Checklist–Military Version; Gray, Litz, Hsu, & Lombardo, 2004). Childhood trauma, neglect, and maltreatment were measured by the 10-item Adverse Childhood Experiences Scale (ACE; Felitti et al., 1998). Responses were summed to calculate total exposure to childhood traumatic events. Among an international sample, the Cronbach's alpha for the ACE was 0.80 (Kazeem, 2015). PTSD symptoms were measured by the 20-item PTSD-Checklist-5 (PCL-5) (Weathers, Litz, Keane, Palmieri, & Schnurr, 2013). Severity of symptoms is endorsed from a 0–4 range, where 4 is the most severe. Psychometric data on the PCL-5 show it to have strong internal (r = 0.94), convergent (rs = 0.74 to 0.85), and discriminant validity (rs = 0.31 to 60; Blevins, Weathers, Davis, Witte, & Domino, 2015).

2.3.3. Substance use & treatment

Past year and lifetime substance use were assessed by the DSM-5 Checklist for Substance Use Disorders (Hudziak et al., 1993). This checklist assessed for age of first use, past year substance use, DSM-5 criteria for all substances used in the past year, and age at which two or more DSM-5 criteria were first experienced. Historical experiences of substance use treatment were assessed by selected questions from the Treatment Services Review (TSR; McLellan, Alterman, Cacciola, Metzger, & O'Brien, 1992) for alcohol/drug and psychiatric treatment. The TSR has been shown to have good reliability and validity at comprehensively assessing treatment history (McLellan et al., 1992; Mclellan et al., 1998; McLellan et al., 1999). Additional questions assessing for the presence of chronic medical conditions (yes/no) and chronic pain (yes/no) were added.

2.3.4. Depression and anxiety

Depression and anxiety were assessed using the Patient Health Questionnaire-9 (PHQ-9; Kroenke, Spitzer, & Williams, 2001) and the generalized anxiety disorder 7-item scale (GAD-7; Spitzer, Kroenke, Williams, & Löwe, 2006), respectively. The PHQ-9 is a 9-item medical screener for Major Depressive Disorder used widely across medical facilities. Psychometric studies show a Cronbach's alpha ranging from 0.83 (Kroenke et al., 2001) to 0.89 (Cameron, Crawford, Lawton, & Reid, 2008). The GAD-7 is a 7-item medical screener for generalized anxiety disorder, also widely used, and shown to have a Cronbach's alpha of 0.92 (Spitzer et al., 2006).

2.4. Data analysis

Qualitative semi-structured interviews were digitally recorded and transcribed verbatim without client or provider identification. Transcripts were uploaded into qualitative data management software, NVivo v11. Two qualitative codebooks were developed from the interview guides, one for clients and one for providers, and modified based on team consensus (TS, LF, MS). Inductive codes could be added to the codebook during coding. Four interviews (two clients and two providers) were coded separately by two team members (TS and LF) and reviewed for consensus. Interrater reliability for these two client interviews was 80.4% and for provider interviews was 62%. Codes were compared and discussed until consensus was reached. One co-author (LF) coded the remaining transcripts, which were then reviewed by a third, independent coder (SK).

A grounded theory approach (Chun Tie, Birks, & Francis, 2019) was applied to identify patterns within the qualitative data. For the two research questions reported in this study, 12 codes from participant transcripts and 9 codes for provider transcripts were assessed. Four independent raters (TS, LF, SK, MS) read excerpts from each code and generated independent themes. Percent agreement for extracted independent themes was 90% for providers and 84% for clients. Themes were then compared and discussed among all authors until consensus was reached; one co-author (DH) acted as an independent observer of the data and helped moderate final consensus on themes. Direct quotes were maintained in the reporting or results to capture participant voices. Last, descriptive statistics (means, standard deviations, frequencies, and percentages) were computed to describe general participant characteristics.

3. Results

3.1. Client characteristics

Table 1 provides demographic characteristics of clients who participated in the interviews. Women were on average 41.91 (SD = 9.38) years old and most had at least a high school education (n = 9, 81.8%), children (n = 9, 81.8%), record of prior arrest (n = 9, 81.8%), received the majority of their housing and economic support from others (n = 9, 81.8%), and used their smart phones every day (n = 11, 100%). Clinically (Table 2), based on self-report data, the women in our sample exhibited mild to moderate symptoms of depression (M = 14.1, SD = 4.25, Range = 8–20; Kroenke et al., 2001) and reported moderate anxiety symptoms (M = 11.0, SD = 5.83, Range = 3–21; Spitzer et al., 2006). With regard to past year substance use, endorsement of opioids was the highest (n = 8, 72.7%), followed by alcohol (n = 6, 54.5%) and cocaine (n = 5, 45.5%). Most women reported first using opioids in early adulthood (M = 23.0, SD = 8.12, Range = 14–37). In addition, women endorsed multiple severe lifetime substance use disorders. While all women in the study reported a lifetime diagnosis of an OUD (100%), ten women (90.9%) reported a severe lifetime OUD as defined by DSM-5 criteria (American Psychiatric Association, 2013). Moreover, four reported a severe alcohol use disorder (36.36%) and three (27.27%) reported a severe cocaine use disorder. The most common medication used to treat OUD was Suboxone (n = 7, 63.6%), followed by methadone (n = 4, 36.4%), and Subutex (n = 2, 18.2%).

Table 1.

Demographic characteristics.

Variables	Clients (N = 11) n (%) or M (SD, Range)
Demographics
Age	41.91 (9.38, 32–57)
Race
Black/African American	4 (36.4%)
White	6 (54.5%)
Hispanic/Latino	1 (9.1%)
Education
Some high school	2 (18.2%)
Completed high school/GED	2 (18.2%)
Some college/Associate's degree	6 (54.5%)
Bachelor's degree	1 (9.1%)
Marital Status
Married	1 (9.1%)
Separated	2 (18.2%)
Never married	8 (72.7%)
Insurance
Medicaid/Medicare	11 (100%)
Children (% yes)	9 (81.8%)
Number of children <18	4 (36.4%)
Foster care/Other caregiver	1 (9.1%)
Employment
Retired/Disabled	5 (45.5%)
Unemployed	6 (54.5%)
Arrested (% yes)	9 (81.8%)
Incarcerated (% yes)	7 (63.6%)
Jail/Prison (months)	31.71 (51.17, 1–144)
Probation/Parole (% yes)	4 (36.4%)
Majority of support from other (% yes)	9 (81.8%)
Smart phone use
Health-content (% yes)	10 (90.9%)
Poor service in past month (days)	4 (8.93, 0–30)
Everyday use of phone (% yes)	11 (100%)

Table 2.

Clinical characteristics.

Variables	Clients (N = 11) n (%) or M (SD, Range)
PHQ-9 Total Score	14.09 (4.25, 8–20)
GAD-7 Total Score	11 (5.83, 3–21)
Past year substance use
Alcohol	6 (54.5%)
Cocaine	5 (45.5%)
Heroin	0 (0%)
Cannabis	2 (18.2%)
Opioid analgesics/Synthetic opioids	8 (72.7%)
Other Drug	2 (18.2%)
Age of first use
Alcohol (n = 11)	15.27 (2.76, 12–21)
Cocaine (n = 10)	17.5 (3.41, 13–24)
Stimulants (n = 5)	24.60 (15.63, 13–52)
Cannabis (n = 11)	15.18 (2.44, 12–30)
Opioids (n = 11)	23 (8.12, 14–37)
Other Drug (n = 2)	16.5 (0.70, 16–17)
Lifetime DSM-5 Criteria (Range: 1–12)
Alcohol	7.60 (4.50, 0–11)
Cocaine	7.25 (4.35, 1–11)
Stimulants	0 (0)
Cannabis	6 (0, 0–6)a
Opioids	9.67 (3.28, 1–11)
Other Drug	5.50 (0.71, 5–6)
ACE sum	4.36 (2.94, 0–8)
LEC-5 sum of “It happened to me”	7.45 (2.66, 3–13)
Accidental traumas	20 (36.4%)
Intentional traumas	34 (51.5%)
Other Traumas	28 (42.4%)
PCL-5 total score	42.45 (17.9, 18–67)
Lifetime treatment episodes
Alcohol abuse	0.36 (0.67, 0–2)
Drug abuse	6.10 (5.40, 1–20)
Current medication assisted treatment
Suboxone	7 (63.6%)
Subutex	2 (18.2%)
Methadone	4 (36.4%)
Chronic medical condition (% yes)	10 (90.9%)
Chronic pain (% yes)	9 (81.8%)
Medication for chronic pain (n = 10; % yes)	1 (11.1%)

Note:

^aEndorsed by one person (n = 1).

Women reported numerous childhood traumatic events (M = 4.36, SD = 2.94) and lifetime traumatic events (M = 7.45, SD = 2.66). Average number of PTSD symptoms reported (M = 42.45, SD = 17.9) was above the clinical cut off of 33 on the PCL-5 (Weathers, Litz, et al., 2013). Seven of 11 women endorsed symptoms above the cut off of 33, suggesting a PTSD diagnosis. Finally, most women also reported chronic pain (n = 9, 81.8%) or a chronic medical condition (n = 10, 90.9%), but only one reported receiving pain medication. The other women in the study did not report receiving any medication to manage pain.

3.2. Primary qualitative themes

Among clients interviewed in the study, two themes emerged as barriers to trauma treatment: (a) lack of trust in providers and (b) self-blame. Six themes emerged among clients regarding technology-based, trauma informed interventions: (a) ambivalence about the usefulness of a technology-based application, (b) desire for it to increase access to services, (c) preference for it to be an adjunct to treatment, (d) interest in it connecting trauma and substance use symptoms, (e) concerns about privacy, and (f) preference for specific technology platforms.

Among providers, one theme emerged in response to barriers to trauma treatment: (a) limited resources. Six themes emerged in response to trauma-informed technology-based interventions: (a) interest in the application, (b) desire for it to be an adjunct to treatment, (c) an application that could manage symptoms and provide medication reminders, (d) increased motivation and reward, (e) confidentiality, and (f) reduced provider time. Table 3 provides a summary of these themes.

Table 3.

Extracted themes among clients and providers.

Women with OUD and PTSD symptoms	Addiction treatment providers
Aim 1: Barriers to trauma treatment
Lack of trust in providers and peers Self-blame for not receiving trauma treatment	Limited resources to provide trauma treatment (e.g., time, access to trainings, staff)
Aim 2: Perceptions on a prospective trauma informed, technology-based intervention
Ambivalence about usefulness Interest to increase access to services Adjunct to in-person treatment Connect trauma and substance use symptoms Remain confidential Preferred technological platforms (e.g., an application, not social media)	Interest in development and applicability Increase motivation and reward for treatment progress Adjunct to in-person treatment Manage symptom tracking and medication reminders Remain confidential, avoid intruding on patient privacy Reduce provider time and burden

Notably, during the analysis of the qualitative data, both clients and providers appeared to vary in their interpretation of the terms “trauma-informed treatment” and “trauma treatment.” This is an important result from this qualitative study, further discussed below.

3.2.1. Clients: Trauma barriers

Two themes emerged in response to structured interview questions assessing potential barriers to trauma treatment or what ideal trauma treatment might look like. The first theme was a difficulty trusting others—both other clients in therapy groups as well as health care providers—to talk about their trauma experiences. One client explicitly said, “No, not really. You know. You know, I just, you know, I just have to trust the person.” Another participant said, “It's almost like if you were to ask me today, do you wanna go to a group about trauma and speak about your experiences to two people or one person, I would just automatically say no. Because, like, don't go there.”

A second theme that emerged among women was the tendency to blame themselves for not seeking additional trauma treatment. As one participant stated, “I feel as though this facility has been here awhile and you know I have utilized it to the best of my ability. Um, now as far as case management, um, I know it's open to us now, so I need to do more things with coming down here and talking to the case manager. You know, um, and utilizing the services as I should.” Many women expressed “should” or “I” statements in response to not receiving trauma treatment. Moreover, some participants were unsure of what “ideal” or additional treatments they could receive.

3.2.2. Clients: Tech-based, trauma-informed application

Clients' initial reaction to a trauma-informed, technology-based application to assist in their treatment was ambivalence. Some participants expressed interest:

So like if you are, you know, if this technology thing you guys are gonna do, you should have something about structure…Because that's what opiate users need, when they are coming off like Suboxone, they need structure. So if you gave them like plans to do every day or something like that, and they could get points or whatever…

Others expressed hesitation, questioning the design of the application, or complete disinterest due to favoring in-person treatment:

I've been pretty good with technology for so many things, and I just feel like a lot of technology has interrupted me, for people to actually in this position seeking help for mental sobriety. It needs to be done with the person. A lot of it is that we're not heard. We don't have someone to listen to us. We don't get feedback. So, yeah, technically, you get some kind of feedback from a computer, but it's not going to be as helpful as sitting down with a counselor and actually talking about things.

When further probed on which technology features might be useful, the second theme that emerged was using technology to increase access to care and others in treatment. For instance, one participant expressed an interest in using technology to help her talk with her providers to address symptoms of panic disorder that were a barrier to attending the community clinic:

I always thought like, sometimes, sitting down for therapy or something, it's nice to just text because that's the way I normally communicate. I don't really talk on the phone. […] Or just like a method of just going back and forth with somebody who's either a therapist. […] Like when I had panic disorder, I couldn't leave my house, so, to come here, it was really hard until I got on like a hefty dose of medications that, to this day, I'm still on. […] If I had something at that time, I think I could have prevented having to be on so much medication initially to be able to get to treatment.

Another participant expressed an interest in an application to assist with using skills in the moment of need:

When I say somebody to talk to, like somebody who can, like, [say],“Why don't you try this?” Or, like, “These are the skills like, you know, let's talk about how you're feeling.” Like, “Why are you so scared to go in front of people?” Like, you know, just kind of like, honestly being able to talk to somebody that will kind of like bring your mind to a place where the conversation is started for treatment, you know, like or like being treated or skills or like, yeah, helping you get through it, you know?

A third theme that emerged was the desire for the technology application to be an adjunct to standard care. Indeed, some participants expressed hesitation at the idea of engaging in a technology application if it resulted in reduced in-person care. When asked if they would replace standard care with an application, many participants expressed something similar to this participant, “No, uh-uh, absolutely not. I like the hustle and bustle of getting here. You know, because it makes me feel like I'm putting effort into my life. I put effort into using so I want to put effort into my recovery.” Another participant said, “I wouldn't replace it because, what if you can't get through on the computer because it's bad service? You can always go in to your regular therapy and see her face to face. So, so long as you keep the same therapy with the same person in door and on the computer.”

Participants also expressed confusion, as well as intrigue, at the possibility of tracking traumatic stress symptoms and connecting how substance use may be related to self-medication of traumatic stress symptoms. Notably, many participants were unsure of this question, asking the interviewer to further explain the relationship between traumatic stress symptoms and substance use cravings. However, upon explanation, this became a fourth theme of strong interest in further understanding the relationship between trauma symptoms and substance use. For example, one client said, “If it can kind of like, you know, what's it called, connect the dots, you know, a little bit. And like maybe, you know, get you thinking.”

A fifth theme that emerged among client responses to a tech-based intervention was the desire for privacy and confidentiality. As one participant said, “I'm a little hesitant because I would have to know every like detail about it, like would my face…would I have to put a picture up of me? Like what kind of information are they going to require? Like my first and last name?” Moreover, when participants were asked what technological platform would be ideal for such a service, anonymity and confidentiality were of importance, “I like texting an app. Social media, as far as that, I think it would be great but as far as like I would want to be anonymous unless it were more of like a one-on-one type of thing or if I chose to, you know.” Finally, in addition to many participants preferring that the technological platform be an application, a few participants also expressed interest in a directory of resources for treatment:

I guess the only thing I could possibly think of is that there is a site or something that, you know, for someone who's starting out trying to get into a program or trying to get resources, that all the information is there and it's actually going to help. […], you look up different detox programs or residential programs and you're on the phone nonstop with all these different places and they don't have anything open, and they don't have anything to—another number or another place or something to help you. So, a lot of people call these places and they get turned away.

3.2.3. Providers: Trauma barriers

In comparison to clients, providers highlighted more barriers to providing trauma treatment or trauma-informed care to women in OUD treatment. The overarching theme was limited resources for both clients and providers. For example, one provider expressed limited childcare resources for mothers:

I mean it's hard because a lot of them, especially if they have children, you know, it's like they really want to be in treatment, um you know and then they run into childcare issues, they can't come today, or they're sick. […] Their recovery takes a backburner because of the kids and stuff.

Comparatively, a couple of providers expressed time and staff limitations as a significant hindrance to providing more comprehensive trauma-informed care:

At my other facility […] I was able to sit with them for an hour. But a lot of the times, it's difficult, 'cause, you know, you have your clients that you have to do vitals and give meds and take care of them and the clinician also has her caseload. So, I mean, sometimes just not enough staff to meet the needs of the client.

Another provider stated:

I don't have enough time to provide the care that they would need. If I wanna address the trauma, I'd like the program to be a little bit more intensive with that. Right, so for me it would be like, yeah, [I] want to make sure I can see you every week, and so say we are doing a half hour session, [but] I'd like to spend an hour, you know. […] 'Cause, dealing with a trauma, the more you can get in and provide the support quicker, I think it helps them be, you know, more successful in staying in recovery, or at least achieving abstinence if that's where they are at, it depends on the client.

A further barrier for providers was having enough time to receive training in the delivery of trauma treatments:

There's no time to do paperwork. I'm lucky if I can choke something down to eat for lunch in five minutes, so it's like where are you going to find the time to, you know […] Yes, I can have someone come in during a staff meeting for an hour and do some type of an in-service, but for me that's not good enough. I'm talking about getting people training, where maybe you are like taking the course even if it's like a three-day course or something like that. The feasibility of that, like having it happen for the employees at work is like slim to none. You have to take the initiative and go and do it on your own. The issue with that is oh, well then can you get the time off of work to do it? And then there's also the financial piece about it to, 'cause now we are talking you could be paying anywhere from $500 to like $2000 depending on what you want to do in terms of trauma certification and things like that and not everyone can afford it.

Although, another provider expressed feeling unequipped to go beyond assessing traumatic history to discussing it further with clients:

It's sort of like well you get all of this information but what do you do with it? […] I think we're really hesitant to even ask what the trauma was. We might ask “oh, have you been physically or sexually abused”, and they'll say yes, and then we might just like move on and talk about something else. So not finding out what actually happened.

Interviewer: Why do you think it's hard to ask?

Provider: Um, I think the biggest thing well, is we're opening up a can of worms, and we're opening up like a Pandora's box, uh, and then you don't know like what to do with that information.

Together, these responses show that providers experience difficulty in providing trauma treatment, even if they desire to, due to resource limitations: childcare, training, time, and staff knowledge and workload.

3.2.4. Providers: Tech-based, trauma-informed application

Akin to interviews with clients, we first assessed providers' initial reactions to a trauma-informed, technology application in their daily treatment centers. Providers expressed interest in an application, especially in the ability to be a platform of care due to barriers to coming to the clinic. One provider also stated that the application could apply psychosocial skills in real-time:

What I've seen to date is basically when they're in the throes of something, somehow having them be able to pull up skills or reminders of skills that they've been taught in terms of distraction and breathing techniques, and if they've been trained in mindfulness techniques in any way.

When further probed about what design features providers may like in the development of such an application, many providers expressed wanting an application that worked as an adjunct to in-person treatment, the second theme that emerged from transcripts. One provider stated:

I mean what you would want is something that you could start in session, and then it would be a reminder through the week, so that they can continue whatever it was that was started. I mean, one of the problems in much therapy, if you're doing weekly therapy, a lot gets lost as soon as they leave the office. And if it doesn't get reinforced the next week, you're sort of starting back where you were the previous week.

In addition, when asked if they would be interested in tracking symptoms to identify triggers to engage in substance use, a third theme arose. Providers expressed interest and provided additional suggestions, such as including medication reminders or tracking clients' moods to clarify emotional triggers that may initiate craving or use:

So like the triggers related to something very specific, because a lot of times, you've got triggers which are more general, like people are often triggered by boredom. They're often triggered by anger. They're often triggered by feeling invalidated. But those seem very like general. It sounds like what you are talking about is something that could provide more specifics about the generalized trigger.

It appeared that the desire to include homework, tracking of symptoms, and medication reminders in the design was rooted in an overall desire to have an application that increased motivation in treatment, the fourth theme. Providers expressed a desire for an application to increase motivation:

And then like you know also having like a reward, like if they're thirty days clean keeping track of that … putting something up like, “Congratulations, you know, you're clean.” Kind of motivating them to continue doing what they're doing and why they're doing it.

Moreover, providers wanted to increase motivation by tracking overall growth in recovery:

Something that marks progress, I think that's really important too. You know like if…so you have an app, and they have skills that they're learning, and then it kind of aggregates the data so they can do a check-in. So since the last time you checked this app, you should have meditated 15 [minutes].

Furthermore, akin to clients, a fifth theme that emerged was the desire for privacy. However, among providers, there was an additional concern to engage in viewing progress in an application for fear that it may intrude on the privacy of clients:

I mean I wouldn't want to…I don't want to invade on their privacy, but maybe just being able to go in and see their trends of how their mood correlated to them wanting to use, and being like, “Hey, you know look at this.” And maybe educating them in terms of trying to get them to see how those two go together.

Finally, many providers were concerned about whether they would have time to incorporate a technology application into their routine practice. Providers expressed that they would only check a prospective technology application among a subset of clients. As one provider put it, they would solely “want a few minutes to just like log in and see progress.” And as another provider stated, “The factor there would be, ‘Where do I find the time to do that?’ Because if…I guarantee you if that had to be off hours, I already do a lot off hours. So if it could be incorporated somehow into my day that would be fantastic.”

4. Discussion

This study explored client and provider perspectives on barriers to trauma treatment and on the development of a trauma-informed, technology-based intervention for women with OUD receiving MAT in community-based substance use treatment settings. Through qualitative interviews, clients and providers shared their current experiences in the delivery and receipt of trauma treatment, and how they could benefit from a trauma-informed, technology-delivered intervention and what attributes may be necessary in the development of such an application. Quantitative results demonstrated that women reported, on average, severe polysubstance use disorders, moderate depression and anxiety, symptoms of PTSD suggesting a full diagnosis, and chronic pain without medicated treatment. Most women were also mothers with adult children, unemployed, dependent on others for housing and financial support, and had a prior arrest history. Emerging themes from qualitative analysis suggested that trauma and trauma treatment were minimally discussed between clients and their respective providers, and secondly, that providers and clients show substantial parity in the design features they seek in a prospective technological application for trauma treatment. Both providers and clients expressed an interest in an adjunct treatment that could implement psychosocial skills in real-world settings, was able to track and monitor triggers and associated cravings, and remained private and confidential. Notably, clients expressed a strong desire for an application to increase overall access to care and connection to providers, whereas providers expressed a concern about having the time to incorporate an application into standard practice.

4.1. Clients

Clients expressed numerous barriers to receiving trauma treatment, including difficulty building trust with providers and peers and taking on sole responsibility for not being able to pursue trauma and OUD treatment. While mistrust of others and self-blame are characteristic of PTSD, these themes also highlight the necessity for providers to initiate conversations around trauma and trauma treatment. Clients were hesitant to discuss traumatic experiences, as is characteristic of PTSD, and thus, providers may need to take the lead to establish rapport and therapeutic alliances that can facilitate discussions about trauma. Indeed, the lack of clarity on what trauma treatment options were available led some participants to blame themselves for not receiving treatment. It is important for clients to seek services and advocate for their respective recoveries, but this may also point to a systemic problem in the delivery of trauma treatment in community facilities. Clients may feel burdened by the research necessary to gain access to trauma-informed care, especially in the presence of co-occurring symptoms of depression and anxiety. Given the onerous work to locate acceptable and effective treatment, community substance use treatment centers, through a technology-based application, could assist in helping clients become aware of trauma treatment options. Such an application could normalize the need for trauma treatment, especially in relation to substance use, and work toward creating a safe environment and instilling hope through visible treatment options.

Clients expressed an interest in the development of a technology-based intervention, but apprehension that use of such an intervention would replace standard, in-person care. This hesitation not only demonstrates the value placed on in-person services, but also the need for a prospective application to work as an adjunct to traditional care. This is a similar concept to many existing technology interventions for substance use disorders (e.g., Beckham et al., 2018; Campbell et al., 2014; Carroll & Kiluk, 2017; Guarino, Acosta, Marsch, Xie, & Aponte-Melendez, 2016). For clients, the primary attraction of a trauma-informed, technology-based application was its ability to increase social connection. Clients were most attracted by the ability of a technology-based intervention to increase access to service providers (e.g., psychiatrists and therapists), as well as female peers in treatment, a feature not available among current trauma and addiction technology-based interventions (c.f. Gilmore et al., 2017). Clients with caregiving responsibilities, transportation barriers, or severe psychological distress were particularly interested in the social aspects a technology intervention could provide. Of note, some women did not explicitly express solely seeking connection with female peers. However, given that most women reported interpersonal violence (51.5% of traumas were interpersonal), histories of substance use with romantic partners, and mistrust of providers, a technological intervention that is both trauma-informed and gender-specific may be warranted.

A notable finding among clients was their lack of knowledge on the relationship between traumatic stress symptoms and substance use. Several clients were not aware that some symptoms, like cravings, are often linked to traumatic experiences, particularly troubling given that women reported multiple adulthood traumas and polysubstance use disorders beyond OUD. When the interviewers provided psychoeducation on this connection, many clients were intrigued and asked for additional information. A technology-based intervention could address this psychoeducation gap with specific content (e.g., Sugarman et al., 2019) that could be accessed repeatedly and on an as-needed basis. This tailored access could support integration of the material and allow clients to utilize the information when needed. Given the widespread reporting of chronic pain among this sample (and individuals with PTSD more broadly), additional content on chronic pain management would be essential. Notably, this type of intervention would need to be developed from the ground up given that the majority of existing trauma and addiction technology interventions do not integrate trauma and addiction treatment (Gilmore et al., 2017).

4.2. Providers

Providers expressed numerous barriers to delivering trauma-related services, including personal discomfort processing trauma with clients, limited time to address trauma, and financial or time constraints to receive training on the delivery of trauma treatment. Previous research has found a similar theme among providers in community substance use treatment facilities: providers express not having the appropriate training or the ability to receive appropriate training to work with populations who have experienced trauma and who have traumatic stress symptoms (Killeen, Back, & Brady, 2015). Indeed, many providers in this study expressed limited resources to address trauma and a generally overburdened case load, which, understandably, limited their mental, emotional, and logistical resources to provide trauma-informed care, even when trained appropriately. A trauma-informed, technological application could be particularly advantageous by relaying pertinent information to clients on trauma and PTSD and creating the context where trauma is normalized and discussing trauma becomes better integrated into care.

Unsurprisingly, given their constraints on time, providers expressed high interest in the use of an application that extended clinician care to reduce their respective workloads. Providers appeared invested in increasing the overall sense of accountability among their clients, eager to have an application that provided rewards and motivation to engage in treatment. However, they also expressed a desire for an application that integrated seamlessly with in-person treatment and required minimal use by the provider. Many providers expressed preference for an application to be viewed in session with a client, underlining the importance of developing an adjunct technology that prioritizes providers' time constraints.

4.3. Trauma-informed care

Given that the two aims of this study were to explore barriers to trauma treatment and characteristics of a potential technological application, the confusion among clients and providers around the definition and application of a trauma-informed care model in addition to what such a model means with respect to screening, assessment, and interventions for trauma within their agencies was notable. The Substance Abuse and Mental Health Services Administration (SAMHSA, 2014) has promoted the need for trauma-informed care, defined as an approach to help create a safe environment, increase awareness of trauma's impact on behavior (e.g., substance use), and apply a strengths-based model that encourages empowerment for the client (Bath, 2008; SAMSHA, 2014). This study's findings suggest that beyond increasing awareness, it remains unclear what actual steps are taken to address trauma for clients receiving services within a community substance use treatment facility.

Several critiques have mentioned that assessment of trauma in and of itself is not synonymous with intervention (e.g., Berliner & Kolko, 2016; Hopper, Bassuk, & Olivet, 2010; Reeves, 2015). In fact, it could be iatrogenic for the clients to be evaluated, identified as a trauma survivor, but then not receive appropriate services. Although routine screening of trauma history was conducted for the clients in this study, it did not appear to be followed by evidence-based trauma treatment approaches. Despite that many evidence-based models have been identified and disseminated for those with comorbid PTSD and substance use disorders (e.g., Concurrent Treatment of PTSD and Substance Use Disorders with Prolonged Exposure [Back, Foa, Killeen, & Mills, 2014] and Seeking Safety [Najavits, 2002]), none was offered to the women in this study, nor were they allowed to make choices about which treatments to receive—a main component of the trauma-informed care approach (Bowen & Murshid, 2015; Hopper et al., 2010; Reeves, 2015). Technology-based interventions could assist providers and clients here. Applications could assess for trauma exposure, provide psychoeducation on the relationship between trauma and substance use, list information on treatment options and selection, and then potentially deliver some components of an evidence-based intervention. This may also attend to the lack of trust with providers expressed by clients; an application may also either flag clients who endorse trauma histories, alerting the provider to raise the issue, taking the onus off the client. These technology-based approaches could help programs struggling to be faithful to a trauma-informed care model with a way to move from simple knowledge and awareness of trauma into providing trauma-specific interventions.

4.4. Prospective technological applications

Themes from this study show promise for the development of a gender-specific, trauma-informed, technology-based intervention. Technology-based interventions are often used as adjuncts to standard care where they can extend the reach of the clinician and deliver content with fidelity. Psychological content may often be diluted over time if delivered by a clinician or not comprehensively delivered, especially if the clinician does not have time. A prospective application for this sample could provide clear, concise definitions of trauma and PTSD symptoms in addition to psychoeducation on the benefit of integrated trauma and addiction treatment. In addition, the autonomy provided by a technology-based intervention might help with expressed client barriers to trauma treatment: trust and self-blame. The confidentiality and opportunity for individualized pacing in a technology-based intervention could allow the client to progress at their own speed in addition to reducing stigma around trauma and PTSD symptoms which, as aforementioned, could possibly empower clients to discuss their trauma experiences in treatment. For instance, allowing the client to work independently to receive initial psychoeducational content on trauma might offer an opportunity to develop initial coping skills to assist in early treatment. Second, to work more collaboratively with providers, real-time tracking of OUD and PTSD symptoms can easily be done using technology-based applications and often with client- and provider-facing dashboards (i.e., where both provider and client can see the same screen). This can allow the client to track progress over time and allow the provider to know when a client might be struggling. An additional consideration is tailoring an application to the needs of women. Providers mentioned that women experience unique challenges to engaging in community substance use treatment (e.g., childcare), and female clients expressed an interest in increased contact with both treatment teams and other women in treatment. Both of these themes suggest that a gender-specific, trauma-informed, technology application may be in concordance with stakeholder preferences. The technology application could not only attend to gender-specific barriers, but also include modules and psychoeducation related to gender-specific triggers and treatment considerations.

The challenge appears, instead, in the ability to implement an intervention that will remain effective in the community long-term, tailored to women, and integrate PTSD and OUD content. The majority of technology-based interventions for PTSD and substance use focus on one disorder rather than both simultaneously (e.g., Sugarman et al., 2019). Moreover, of the two interventions that have been developed to focus on trauma and substance use simultaneously, both have focused on mass trauma (Gilmore et al., 2017). The women in our sample primarily reported interpersonal traumas, suggesting the need for an application that adequately addresses different types of traumatic experiences. For instance, the application may need content on relationship skills or learning safety signals in interpersonal interactions. Additionally, a particular hurdle in delivering trauma-informed, technological interventions is client engagement (Yeager & Benight, 2018). Trauma survivors may be particularly hard to engage with a technological treatment due to high levels of avoidance. For this reason, incorporating a technological application with standard in-person care may facilitate both treatment recovery, provider burden, and client treatment engagement.

4.5. Limitations

A qualitative research design was the most useful approach for this current study, as we were interested in understanding the experiences of provider and client availability of trauma treatment and their interest in and preferences for trauma-informed technology-based interventions. Although qualitative research does not require generalizable samples, findings from this study may reflect unique aspects of the geographic region or treatment programs in which participants were recruited. Additionally, we did not conduct any post-interview respondent checks for validation of results, and our recording method could have influenced responses. In addition, researcher bias is always possible in this kind of study design, given the use of a single interviewer as the instrument for qualitative interviews, as well as because the data analysis was completed by the same individuals who designed the study. However, multiple coders assessed the qualitative data, and thus attempts were made to eliminate bias as much as possible.

5. Conclusions

This study is the first, to our knowledge, to interview stakeholders in the community—both women in treatment and providers—to understand their experiences of trauma treatment and their perspectives on the development of a trauma-informed, technology-based application for women on medications for OUD. Identifying the desired attributes of a technological adjunct is paramount given the rising rates of opioid overdose among women as well as empirical research supporting and national calls for the development of gender-specific interventions for trauma and OUD treatment (Ecker & Hundt, 2018; Hemsing et al., 2016; Marsh et al., 2018; Meyer et al., 2019; Office on Women's Health, 2017; Terplan et al., 2015). Findings suggest that women in treatment for OUD in community facilities frequently experience co-occurring psychological and substance use disorders beyond PTSD and OUD. Despite that, none of the women in our study reported receiving an evidence-based trauma treatment. This is an important gap in the delivery of patient care given the high occurrence of PTSD and OUD and the severity of OUD among women with co-occurring trauma histories and/or traumatic stress symptoms.

This study is a formative step toward future research assessing the acceptability and feasibility of technology-based, trauma-informed interventions. Themes from qualitative interviews suggest that both women in treatment and providers are interested in technological application as an adjunctive treatment to standard, in-person treatment. Ideally, such an application would provide psychoeducation, extend the reach of treatment, and increase access to provider and female peer support. Creating such a user-friendly, psychologically effective, and feasible application addressing trauma and addiction will require an interdisciplinary effort among the fields of behavioral science, computer science, digital science, and business (Yeager & Benight, 2018). However, through developing such an application, multiple gaps in patient care could be addressed. A technology application could increase treatment access for clients, reduce the burden on providers, and address systemic barriers in the delivery of trauma-informed care, through the delivery of knowledge and skills with fidelity. Using mixed methods and involving stakeholders in the development and refinement of such an intervention will be critical to the downstream success of a technology-based intervention.