Journal of Addictive Behaviors,Therapy & RehabilitationISSN: 2324-9005

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, J Addict Behav Ther Rehabil Vol: 2 Issue: 3

Initial Pilot Test of a Group- Texting Intervention to Sustain Opioid Abstinence Following Residential Detoxification and Treatment

Angela L Stotts1*, Thomas F Northrup1 and William D Norwood2
1Department of Family and Community Medicine & Psychiatry and Behavioral Science, University of Texas Medical School at Houston, Houston, Texas 77030, USA
2Department of Clinical, Health, and Applied Sciences, University of Houston – Clear Lake, Houston, Texas 77058, USA
Corresponding author : Angela L Stotts
Department of Family and Community Medicine, University of Texas Medical School at Houston, 6431 Fannin, JJL 324, Houston, TX, 77030, USA
Tel: (713)500-7590; Fax: (713)500-7598
E-mail: [email protected]
Received: March 27, 2013 Accepted: August 22, 2013 Published: August 24, 2013
Citation: Stotts AL, Northrup TF, Norwood WD (2013) Initial Pilot Test of a Group-Texting Intervention to Sustain Opioid Abstinence Following Residential Detoxification and Treatment. J Addict Behav Ther Rehabil 2:3. doi:10.4172/2324-9005.1000110


Initial Pilot Test of a Group- Texting Intervention to Sustain Opioid Abstinence Following Residential Detoxification and Treatment

Relapse to opioids is common among detoxified patients following discharge from residential treatment. Mobile phone, group-texting technology offers an innovative method for delivering effective, lowcost aftercare during this critical period. Recent research suggests acceptance and mindfulness processes might be useful targets in opioid detoxification. An Acceptance and Commitment Therapy (ACT) group-texting intervention was developed and tested in a small pilot study (N=10) with detoxified opioid patients in residential treatment (RT). Participants received 6 ACT-based group sessions while in RT and 4 weeks of ACT group-text messaging following discharge. Overall, the intervention was well received and feasible. Half of the participants received 5 or 6 RT sessions and replied to post-discharge texting prompts. Texting content was generally ACT-consistent, however, the frequency of texts was lower than expected. Two participants sent fewer than 10 texts, and three participants texted between 11 and 20 times during the monthlong intervention. Focus groups indicated acceptability and recommended adjustments to the intervention. ACT mechanisms changed in theoretically predicted ways: Experiential avoidance related to drug and other cues was reduced, and both acceptance and values-driven action were increased across treatment. Preliminary data suggest promise for the novel ACT-based grouptexting intervention, which ultimately, with further development, may be a successful adjunct to aftercare treatment for recently detoxified opioid dependent patients.

Keywords: opioid dependence, opioid detoxification, text messaging, ACT, acceptance and commitment therapy


Opioid dependence; Opioid detoxification; Text messaging; ACT; Acceptance and commitment therapy


The burden of illicit opioid use is substantial. Morbidity and mortality rates are high, and the costs associated with health care, law enforcement, family dysfunction, and lost productivity are significant [1,2]. Although methadone and buprenorphine maintenance continue to be the most effective treatments for opioid dependence [3,4], these approaches remain controversial in the community and increasingly both patients and treatment providers are choosing to forgo maintenance medications in favor of non-pharmacological approaches [5-7].
Unfortunately, opioid detoxification, particularly on an outpatient basis, is associated with high rates of treatment drop-out and illicit opioid use [8-10]. Relapse rates in the six months following inpatient or residential detoxification are high, ranging from 70-–87% [8,11]. And almost all lapses occur within the first month following discharge [12]. For example, Chutuape et al. [8] reported that over 80% of participants had used opioids in the first month after inpatient detoxification. Active participation in intensive aftercare, however, has been found to improve treatment outcome, with attendance in aftercare differentiating those who succeed from those who fail [13-15].
While associated with success, intensive aftercare is timeconsuming, inconvenient for patients, and expensive, and though patients typically voice a desire to participate in post-detoxofication aftercare, rates of actual participation are often low [16,17]. Only about a quarter of those transitioning from inpatient treatment present to after care [18]. Modern mobile phone technology offers an exciting and possibly revolutionary method for delivering effective, low-cost aftercare treatments, and improving engagement in some type of aftercare during the critical period immediately following detoxification and discharge from residential treatment.
Text messaging via the mobile phone as a mode of substance abuse treatment delivery is only beginning to be exploited. The technology is highly portable and ubiquitous as 80-90% of people in the US have text-enabled mobile phones [19]. Text messaging may allow for the extension of interventions into patients’ natural environments, where drug use antecedents such as drug cravings, struggles with unwanted thoughts and emotions, and environmental cues directly challenge the skills taught in treatment. The effectiveness of most behavioural interventions depends on skill acquisition followed by generalization to the natural environment, requiring practice of skills in vivo. Currently, our evidence-based therapies include no consistently effective means to ensure that this happens, nor to monitor and potentially intervene during high-risk, crisis periods. Mobile phone technology may offer a method to monitor patient progress and track skill practice and effectiveness in natural environments, potentially over long periods of time at relatively low cost.
Research investigating the use of mobile phones in the treatment of substance dependence is scarce, yet promising [20]. Mobile phone text-messaging interventions have most often been tested with smokers, with quit rates ranging from 22-43% [21-24]. Most studies, however, have combined texting with internet or video treatment components requiring computer or expensive telephone technology [20], which is currently inaccessible by many lower income substance abusers. To date, all text messaging interventions have been delivered individually.
Substance abuse treatment of drugs other than tobacco, however, has traditionally sought to harness the benefits of peer group interaction in the service of maintaining abstinence. The technology is now available to allow dynamic, real-time group interaction via text messaging (group-texting) that can be moderated by a group leader or therapist. Interacting with peers under the guidance of a treatment provider conforms directly to the community treatment experience in which group therapy is the primary method of treatment delivery, and this technology has the potential to extend the therapeutic benefits of peer interaction into the natural environment during the critical aftercare period. Group-texting has become a familiar and commonplace mode of communication for many cell phone users and may offer the benefits of a therapeutic group, standardized theoretically-driven messages, in addition to the convenience, ease and quick feedback afforded by texting interventions. This technology may be ideally suited to the needs of opioid-dependent patients following detoxification.
Cognitive-behavioural theory and treatment is the natural choice on which to base a text messaging intervention. Few behavioural treatments, however, have been designed to address the unique psychological experiences and somatic symptoms specific to opioid detoxification, such as an overreaction to mild physical symptoms and fear of a drug-free state (including anxious arousal and unwanted thoughts) [25], which increase the probability of negative outcomes. Recent theory and research suggest that interventions engaging acceptance and mindfulness processes might be especially effective in providing adaptive, alternative responses to such difficult internal experiences as unpleasant withdrawal symptoms [26] and unwanted thoughts, sensations, and emotions (e.g., shame, negative selfevaluations, anxious arousal, and hopelessness) [27].
Acceptance and Commitment Therapy (ACT) targets both acceptance and mindfulness processes and has demonstrated efficacy with other substance abuse populations [27-30]. SAMHSA has listed ACT on their National Registry of Evidence-based Programs and Practices [31].
The theory underlying ACT maintains that a great deal of suffering emerges from persistent attempts to control or escape various aspects of experience (i.e., thoughts, emotions, bodily sensations), which in ACT is referred to as experiential avoidance [32,33]. The struggle to eliminate or control suffering or adversity removes the individual from the present moment and interferes with the ability to choose behaviours consistent with his or her goals. ACT helps clients relinquish the struggle, allowing actual experience rather than feelings and thoughts (e.g., I can’t stand this) to direct behavior. The primary goal is not to lessen distressing thoughts, emotions, or physical sensations but rather to decrease their impact. This is accomplished by increasing willingness and openness to experience distress in the service of some broader goal or value. This in turn will facilitate a broad and flexible behavior repertoire in the face of typically avoided private experiences.
We recently reported on the first study of ACT specifically targeting opioid detoxification [26]. This 24-session ACT protocol, in general, sought to decrease experiential avoidance by increasing willingness and acceptance of the distressing physical and psychological experiences (e.g., fear of detoxification, withdrawal symptoms, fear of relapse). Patients were assisted with noticing their internal triggers, abandoning their attempts to manage these triggers with active avoidance or suppression (e.g., drug use), and making commitments to engage in behaviours consistent with their chosen values or goals (rather than allowing negative thoughts, feelings, and physical withdrawal symptoms to determine behavior). The data from this initial study are promising [26]. Opioid dependent patients presenting for methadone detoxification were randomized to receive individual therapy sessions of either ACT or Drug Counselling (DC). By the end of treatment, the number of participants in the ACT condition who were successfully detoxified was nearly doubled compared to those in DC. ACT may be an especially promising treatment upon which to base a text-messaging intervention targeting opioid detoxification.
Initial development of an ACT-based group-texting module presented many challenges, however. For example, it was unclear whether or not patients would find texting, and in particular ACTbased group-texting, to be an acceptable mode of treatment delivery (e.g., would patients participate? Would they be willing to use their phones for this purpose? Could coherent and useful ACT concepts be conveyed in brief text messages? Would patients read the ACTbased messages and act on prompts to engage in group discussion? Would they report via text on their own behavior, such as skills practice?). We also feared the worst: those patients might “hi-jack” the system and use the group-texting for inappropriate or illegal purposes. Additional unknowns included the number of texts per day or per week that might be optimal (e.g., enough to engage patients without being experienced as burdensome), the role of the counsellor facilitator, and whether research-treatment staff could monitor group-texting interactions with sufficient ease. With this in mind, we conducted a small pilot investigation to evaluate the feasibility and acceptability of delivering a group-texting aftercare intervention following opioid detoxification in a residential treatment setting.


The ACT group-texting (ACT-GT) intervention, designed to begin during residential treatment immediately following detoxification had two components: (1) Group therapy while in residential treatment (RT) to provide a foundation of ACT concepts; and (2) Group-texting after discharge with regularly scheduled ACT-based prompts to promote continued engagement with the therapeutic processes during the aftercare period.
RT ACT groups
The protocol was largely drawn from the first author’s methadone detoxification ACT manual [26,34] and previous substance abuse ACT protocols [29,30], with modification of contents to reflect issues specific to the post-detoxification period. The RT ACT protocol consisted of six, 60-minute, group sessions delivered thrice weekly for two weeks following detoxification. Emphasis was placed on identifying personal values and viewing opioid detoxification and abstinence within the context of valued life directions, noticing internal triggers (e.g., craving, anxiety, negative effect, withdrawal experiences, etc.), abandoning attempts to manage triggers via active avoidance, suppression or other control-based strategies (e.g., substance use), being open to experiencing these triggers and other psychological difficulties without defense (i.e., acceptance), and making commitments to actions consistent with client-chosen values. Additionally, the ACT protocol was designed to help clients learn to attend to the present moment, to mindfully experience distressing thoughts and feelings from an observer perspective, and notice becoming “entangled” by them.
Each session started with a 5 to 10-minute mindfulness exercise (e.g., focusing on breathing, sitting still, noticing and observing thoughts, feelings, bodily sensations), followed by a review of clients between-sessions activities. Session-specific components were introduced using clients’ personal experiences related to detoxification, abstinence, and/or other life events. Consistent with ACT, sessions comprised metaphors and experiential exercises to compliment didactic discussions.
ACT-Based group-text messaging
The group-texting phase of the intervention began immediately upon RT discharge. ACT-related “prompts” were sent 3 times a week at various times of the day for one month, encouraging the use of mindfulness and acceptance skills in real-time. Authors, along with the counselor facilitator, developed the prompts based on metaphors and exercises that resonated most with participants during the RT ACT groups and that targeted the ACT theoretical processes, such as willingness to experience distressing thoughts and feelings, mindfulness, and values-based action (Table 1 for a list of texting prompts). Client input was solicited about which concepts would be most helpful to see in text form after discharge. During the texting phase, participants were encouraged to reply to the prompts as well as reply to each other in order to benefit from the group support aspect of the intervention.
Table 1: ACT-Based texting prompts.
All text messages were administered via the BIGTxt web-based system that was designed at the University of Texas Houston Health Science Center’s School of Biomedical Informatics to distribute instant or scheduled individual or group-text messages related to patient health. BIGTxt has multiple layers of security, which enabled secure and confidential group-text delivery by trained study staff. All mobile phone numbers were entered into BIGTxt and all participants (and counselor facilitators) were part of the same texting group. Messages initiated from BIGTxt and by participants or facilitators occurred by sending the text messages through a non-identifying centralized phone number. All members of the group received every message that was sent from BIGTxt or by another group member via this number. Specifically, BIGTxt interfaces with a third-party short-message system application (i.e., Twilio) to distribute messages through the centralized phone line. This way, individual participant phone numbers were not visible to other group members. Only study staff authorized to use BIGTxt had access to individual phone numbers. The BIGTxt database of users and messages retained all content for secure viewing (e.g., network password protection). Study staff monitored the texting content in real-time to ensure appropriate usage. BIGTxt allowed 160-character messages and the ACT-related prompts were tailored to fit within this limit.


This one-arm pilot study was approved by the University of Texas Medical School at Houston Committee for the Protection of Human Subjects (HSC-MS-12-0458) and was conducted in accord with the Helsinki Declaration of 1975.
Participants and procedures
Admissions to a residential substance abuse treatment facility in Houston, TX during a one-month period were screened by intake staff for opioid use. Participants willing to be approached by research staff gave written permission for screening by Research Associates (RA) and were screened toward the end of their detoxification program. Eligible participants were male and female adults (>18 years of age) who met criteria for opioid dependence and were currently undergoing detoxification. Participants also: 1) reported opioids as the primary drug of abuse; 2) reported opioids were used illicitly [heroin or oral opioids without a valid prescription or using more than prescribed]; 3) were in generally good physical and psychiatric health except for possible acute drug-related problems; 4) were able to provide the name of at least one person who could generally locate their whereabouts; and 5) were willing to use their personal mobilephone text messaging plan. Participants were ineligible if any of the following conditions were met: 1) severe cognitive, and/or psychiatric impairment [per judgment of residential treatment provider and research staff] that precluded cooperation with study protocol; 2) chronic pain or other chronic medical condition for which ongoing opioid therapy was prescribed; 3) inability to read, write, or speak English, or; 4) impending incarceration. Within the one-month period, 18 patients were approached, 15 were screened, and 10 were enrolled. Distance from home to the treatment facility, lack of interest in research, and failure to meet opioid dependence criteria excluded 8 patients.
Participants who met the inclusion criteria completed the informed consent process. Pre-treatment assessments were completed following the consent. Participants completed research visits once per week while in residential treatment and during the one month texting phase after discharge. One month after the end of the group-texting intervention, participants completed a 1-month follow-up visit (i.e., 2 months post-discharge). A focus group was conducted by the first and second author after the 1-month follow- up visits were completed. Research visits took between 30 and 60 minutes to complete. Participants received no monetary incentives for participation in this unfunded pilot study.
Residential-based ACT groups: The ACT group therapy was conducted by a masters-level therapist who had worked in the field of substance use treatment for several years and had previous ACT training by the third author. Three cohorts of 3-4 participants each were conducted, each cohort received the 6 ACT-based groups (conducted 3 times per week for two weeks) lasting approximately 60 minutes each. Identical ACT principles were covered in each cohort. The groups were considered elective groups and conducted within the context of an abstinence-based residential treatment program. In this program, agonist medication (e.g., buprenorphine) is only used as needed to assist detoxification but not continued in residential treatment. Non-opioid medications to treat symptoms of withdrawal are provided as needed. The majority of treatment is group-based and the length of stay is typically between 7 -– 30 days, with an average of 14 days for opioid dependent patients.
Monitoring and supervision: All group sessions were audiotaped. The first author listened to each session and provided feedback and guidance to the counselor regarding the subsequent session. The counselor drafted a detailed summary of group topics and discussions to inform further treatment development. Rigid manual adherence was not required in this initial pilot study which allowed the counselor to select the most applicable ACT strategies and techniques. The same core ACT therapeutic processes were targeted in each cohort, however.
Group-text messaging with ACT: After discharge from RT, participants entered the 1-month, group-texting intervention. Thrice weekly ACT-based messages were delivered at various times throughout the week. Messages were delivered to participant phones as early as 7:30 A.M. and no later than 5 P.M. to allow participants sufficient time to view and respond to the texts before going to sleep.
The counsellor who conducted the residential groups monitored and facilitated the group-texting. Participants were encouraged to initiate or respond to the texting prompts as well as messages from one another. Guidelines for appropriate and safe texting practices were reviewed with the group prior to the start of this phase. Specifically, group members were asked to avoid: 1) texting while driving or operating machinery; 2) discussing sexual matters; 3) texting after 11 P.M. at night [when facilitators might be unable to monitor]; 4) using profanity; 5) soliciting or offering money or drugs from/to other group members; 6) making threats of any kind; 7) and sending texts about physical or psychological emergencies (e.g., suicidal ideation). Participants were given a set of safety procedures to follow in the event of physical or psychological distress. Participants were made aware that participation in the texting portion of the intervention was contingent upon their following these guidelines.
Focus group: Participants who had completed both phases of the project (RT groups and group-texting after discharge) participated in a focus group to discuss their experiences during the intervention. Three attended the focus group and the fourth provided feedback to the first author by phone. The focus group lasted approximately 45 minutes and participants were queried about all aspects of the intervention, including the acceptability of residential groups (particularly the ACT-based content), the ease and acceptability of using a mobile phone to participate in the text-based group, the adequacy of ACT-based message conveyance, and suggestions or ideas for improvements. Both focus group facilitators transcribed participant responses and themes were compared after the meeting adjourned.
Measurement time points were at pre-treatment, weekly during treatment, and at 1-month post-treatment. The baseline assessment was conducted at the end of or immediately following opioid detoxification and prior to any research treatment. The 1-month follow-up was conducted approximately 2 months following RT discharge.
Opioid and other substance use: Drug Use History was measured using the Addiction Severity Index, Lite [35]. Reported drug use was assessed using Timeline Followback methodology. Per established methods [36], Timeline Followback interviewing was used to obtain continuous data regarding opioid use in the 30 days prior to detoxification up to the day before the last study visit. A qualitative urine drug screen was conducted in the last week of the texting intervention and at the 1-month follow-up.
ACT mechanisms: acceptance/avoidance & action/inaction: The Acceptance and Action Questionnaire-9 AAQ-9: [32] measured general level of experiential avoidance related to internal experiences. The AAQ-9 is a 7-point Likert-type scale with adequate reliability, alpha of 0.72 - 0.79 [32], that asks about avoidance of emotions, believability of thoughts/emotions, and the inability to act consistently with ones values in the presence of difficult thoughts and feelings. Higher scores indicate higher levels of experiential avoidance. The AAQ-9 comprises two subscales, Acceptance and Action, which are scored in the opposite direction from the full scale. High scores on the Acceptance subscale indicate a willingness to experience unpleasant internal thoughts, feelings, and bodily sensations. High scores on the Action subscale indicate an increased engagement in values-directed behaviours despite unpleasant internal experiences.
The Avoidance and Inflexibility Scale-Substance Use version AIS [28] is a 13-item measure designed to evaluate endorsement of avoidance strategies and inflexibility in behavior specifically related to drug use and withdrawal. Greater scores on this Likert-type scale indicate a more avoidant and inflexible stance toward unpleasant internal experiences related to drug use.
Treatment evaluation scales: An author-developed Treatment Evaluation Form (TEF) was administered to capture the acceptability of the intervention and the satisfaction of the participant for the ACT groups and post-discharge texting portion. Responses ranged from “1 = Strongly Disagree” to “5 = Strongly Agree,” with higher scores indicating greater acceptability ratings from participants.
Attendance, retention and focus group data are presented descriptively. ACT mechanisms were analyzed using pairedsample Student’s t-tests with pre-treatment and 1-month follow-up measurements (approximately 2 months after RT discharge; N=4; SAS, version 9.3). All significance tests were evaluated at the p < 0.05 level.


Consented participants (N=10) were mostly from the young adult program and male (n = 1 woman), with a mean age of 25.1 (SD=3.6). All were Caucasian, as is typical of opioid users in Houston. Most were abusing prescription opiates (n=6), with 3 using heroin and 1 abusing methadone. Days of opioid use out of the last 30 days (prior to residential admission) averaged 21.8 days (SD=9.9 days). 70% of participants at pre-treatment endorsed at least some HIV-risk sexual practices with 40% endorsing high-risk sexual behaviours and 20% high-risk injecting practices, and 10% reporting both. Average length of time in RT treatment was 18 days.
Attendance and retention
On average, participants attended 4 RT group sessions. Two participants attended zero sessions and lost contact with study staff for the remainder of the study; 1 participant attended 1 session; 2 participants attended 3 sessions; and 5 participants attended 5 or 6 sessions. One participant moved out-of-state severing communication, 1 re-entered residential treatment at another facility and fell out of communication with study staff, and one was lost to follow-up. Post-discharge follow-up data was collected on 4 participants. The lack of incentives for returning to the treatment center for research visits undoubtedly affected retention.
Feasibility of texting as treatment
With regard to the group-text messaging component, 5 of the 8 participants who attended at least one or more RT group replied to at least one post-discharge texting prompt. Two participants sent fewer than 10 texts, and three participants texted between 11 and 20 times during the month-long intervention. Of the 56 total texts sent, 38% were ACT-consistent, 4% were ACT inconsistent, 38% were general status updates, and 18% were greetings or other non-relevant content. Coding was done by consensus by the first and second authors. Thus, overall, texting content was ACT-consistent; however, the frequency of texts was lower than expected.
Focus group feedback & treatment evaluation
Focus group participants communicated high levels of enthusiasm for the ACT-based content they received during their RT ACT groups. One participant felt the groups were “liberating,” while another commented that he liked the idea of “thoughts are just thoughts.” Average treatment satisfaction ratings on the 6-session ACT group ranged from 4.3 to 4.8 (possible range: 1-5). Participants uniformly endorsed feeling engaged (M=4.8), finding the exercises helpful to recovery (M=4.8), being willing to recommend this program to others (M=4.5), and having new ways to approach recovery (M=4.8).
Focus group participants conveyed that the ACT-based texting component was convenient and an appropriate medium for an aftercare intervention; one participant commented, “You know how us addicts like instant gratification.” Furthermore, the participants reported that the ability to receive immediate input and examples for how others used the skills was very helpful. The participants also noted several areas for potential improvement. Participants appreciated the “thought-provoking prompts,” and acknowledged thinking about the prompts often throughout the day. While the participants all found the text reminders increased attention to ACTconsistent ideas, most agreed that they were hesitant to reply with their reactions unless they were explicitly prompted or encouraged to do so. Every participant reported that they read and reflected on each text; however, responding was less frequent than expected. It was rare for a participant to initiate a text, especially on days when no prompt was sent by the study team from BIGTxt. One participant suggested that future approaches incorporate action-based prompts (or “exercises”) that asked for follow-up reports on the activity. Further, assigning exercises or reading in a workbook via text was suggested by several participants as a resource to refresh ACT knowledge and to provide supplemental information that may not have been covered in the RT group sessions. When queried, all participants acknowledged that incentivizing the number of text responses would have increased the overall response rate. Average evaluation ratings on the 1-month group texting component ranged from 3.5 to 4.8 (possible range: 1-5). Participants reported that they appreciated receiving the mindfulness reminders and peer support (M=4.3), would recommend the texting component to others (M=4.3), and that they found the convenience of texting appealing (M=4.0).
Focus group participants shared other ideas for future modifications as well. The participants expressed that 3 texting prompts a week were too few and many participants suggested sending them daily or more often (i.e., multiple prompts each day of the week). Participants felt the counsellor facilitator should have been more involved to facilitate discussion. Several participants expressed a desire for one-on-one communication with the counsellor facilitator when they felt uncomfortable sharing particular topics with the group. Related to the area of comfort with other group members, focus group participants shared that it would have been ideal to have an introductory message for each person who entered the texting group (once they were discharged) in case they were unfamiliar with the person or had not overlapped with them during residential treatment.
Opioid use outcomes
Three out of the four participants who completed both the RT group and group-texting phases of the intervention reported sustained opioid abstinence after discharge per TLFB interviews (verified by UDS at post-discharge week 4 and the 1-month followup). One participant reported a 2-day opioid lapse that occurred midway between the end of the group-texting phase and the 1-month follow-up; following this lapse he was abstinent for the remainder of the study.
ACT Mechanisms
Examination of change from pre-treatment to the 1-month follow-up indicated decreased experiential avoidance and greater acceptance of internal experiences as measured by the AIS (t=5.87, df=3, p<0.01) and the AAQ-9 (t=6.19, df=3, p<0.01) and its Acceptance (t= -3.39, df=3, p<0.05) and Action (t= -10.46, df=3, p<0.01) subscales. AAQ-9 scores at pre-treatment (M=45) fell within the upper quartile AAQ-9 > 4132 among clinical populations. Scores at the follow-up were slightly below the mean of Caucasian, nonclinical populations (M=32.9). Data are graphed individually from each of the 4 participants (Figure 1).
Figure 1: Avoidance & Inflexibility Scale (AIS) and Acceptance & Action Questionnaire, 9-item version (AAQ-9; with subscales) scores from pretreatment (baseline) to 1-month follow-up. Only participants (N=4) who completed a 1-month follow-up assessment were graphed. Greater scores on the AIS and AAQ-9 indicate higher levels of avoidance and lower levels of acceptance and action, respectively. Greater scores on the Acceptance and Action subscales of the AAQ indicate higher levels of acceptance and action.


Technology-based interventions have yet to be integrated into substance abuse treatment programs. An innovative, theoreticallydriven, group-text messaging intervention was developed and tested with recently detoxified opioid dependent subjects in residential treatment to ascertain feasibility, acceptance and preliminary data on mechanisms and opioid use outcomes. While the number of subjects was few, the data and experience gained is critical to the development of such an intervention. Focus group data were highly informative. Participants were very interested and pleased with the chosen theoretical perspective, ACT, and uniformly found the experiential exercises and metaphors to be applicable and useful. The grouptexting component was well-received, extended the intervention, and was highly relevant, particularly for the young adult population. The data from participants who completed treatment and the one-month follow-up indicated that from pre-treatment to one-month posttreatment, the ACT mechanisms changed in theoretically predicted ways. Experiential avoidance more generally and in relation to drugrelated cues was reduced, and both acceptance and values-driven action were increased.
Mobile phones are inexpensive and ubiquitous even in lower income populations [19]. Many people have their phones with them every hour of every day, and the majority of mobile phone users use text messaging to communicate with others, making this a natural technological avenue to pursue for intervention. Individual text messaging has been found effective in several smoking studies [21,23,24] and no differential adverse events were found with regard to texting and driving between texting and non-texting groups [37]. To our knowledge, no one has developed or tested a grouptexting intervention in any substance abusing population. Grouptexting conforms directly to the community treatment experience in which group therapy is the primary delivery method and offers larger potential benefits relative to individual texting. Groups can be powerful agents of change, encouraging peer support and pressure to abstain from drugs; a reduced sense of isolation; observation of adaptive coping by others; feedback concerning the values of other group members; and coaching, support, and reinforcement for undertaking difficult or anxiety-provoking tasks. We chose to target recently detoxified opioid dependent patients based on our past work with ACT and opioid detoxification, however, developing a platform treatment using texting technology may benefit many types of substance abusers for whom relapse is the norm and the recovery trajectory is not linear. Group-texting can both extend treatment into patients’ natural environment and offer a peer-based recovery support system.
More developmental work is needed, however, for this intervention to reach its full potential. For example, texts are limited in number of characters, making it difficult to present new information. Participants recommended linking the text messages with additional supplemental material that included educational information and exercises to increase their acceptance and mindfulness skills and values-driven action. Also, the role of the counsellor needs to be better defined. Increasing counsellor participation decreases the efficiency and increases the cost of the intervention, but perhaps not to a large extent. It may be prudent and more effective to have a direct mobile phone line to the counsellor for times during which participants feel that relapse is imminent and wish to speak with someone confidentially. Finally, our participants were often reticent to text even if they felt they had something to contribute, although it was not clear whether this was due to evaluation anxiety or not wanting to put forth the effort or for some other reason. Incentives for initiating texting may be needed and could be faded once peer texting group relationships have been established. In sum, the next step in the iterative developmental process will involve changes related to supplemental materials, direct counsellor access, and incentives to increase texting.
Limitations of this small pilot study are noted. As the study was without funding for compensation and incentives, it was challenging to get participants to return to the treatment center for research visits. Thus, the outcome and mechanism data were gathered from a self-selected sample of opioid-users who were likely more motivated for treatment or who resonated more fully with the treatment. Focus group data may reflect bias as well. Participants who dropped out of treatment may have had less favourable evaluations, different recommendations for improvement, and poorer outcomes. Participants on which we collected focus group and follow-up data had higher levels of treatment exposure, however, perhaps making their data more relevant.
Overall, we believe that innovative methods of treatment and treatment delivery are needed to improve opioid detoxification outcomes and, in turn, to decrease the substantial health and social costs associated with chronic opioid dependence. Our preliminary data suggest promise for the novel ACT-based group-texting intervention, which ultimately, with further development, may be a successful adjunct to aftercare treatment for recently detoxified opioid dependent patients. Effective interventions via mobile phones in the aftercare period following residential treatment discharge would be contemporary, efficient, and cost-effective. Continued research and development are needed.


The authors wish to thank Amber Thomas-Gordon and Swetha Mulpur for recruiting participants and collecting data for this study. Memorial Hermann, Prevention and Recovery Center (MH-PaRC) staff helped immensely in the conduct of this study. Specifically, Paige Harris skillfully conducted the ACT groups and monitored the text portion of the study, and Matthew Feehery, CEO, and Ruthann Adam, Clinical Program Director, provided numerous resources and immeasurable support. We also wish to acknowledge the University of Texas Houston Health Science Center’s School of Biomedical Informatics who managed the BIGTxt system.


Track Your Manuscript

Recommended Journals