This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
Group-based formats typically used in low-resource substance use disorder (SUD) treatment settings result in little individual attention to help reinforce and guide skill use, which may contribute to poor posttreatment outcomes. Smartphone apps offer a convenient, user-friendly, and cost-effective tool that can extend the reach of effective SUD treatments. A smartphone app was developed and integrated into a group-based, brief behavioral activation (BA) treatment for SUD to increase engagement in treatment skills outside clinician-administered sessions.
This study aims to describe the features of the app and its use and integration into treatment, report the participants’ self-reported feasibility and acceptability of the app, and discuss challenges and provide recommendations for future smartphone app integration into behavioral treatments for SUD.
A total of 56 individuals recruited from intensive outpatient SUD treatment received a smartphone-enhanced BA treatment, the Life Enhancement Treatment for Substance Use. Self-reported weekly app use and reasons for nonuse were assessed at posttreatment and at 1- and 3-month follow-ups. In addition, 2-tailed
Participant feedback suggested that the integration of the smartphone app into the Life Enhancement Treatment for Substance Use was feasible and well accepted, and participants found the app useful for planning value-based activities outside of sessions. Self-reported app engagement decreased over the follow-up period: 72% (39/54) of participants reported using the app at posttreatment, decreasing to 69% (37/54) at the 1-month follow-up and 37% (20/54) at the 3-month follow-up. Participants reported forgetting to use the app as a primary reason for nonuse.
This study provides support for the feasibility and acceptability of smartphone-enhanced BA treatment, offering promise for future research testing the integration of technology into SUD treatment. Design decisions may help streamline smartphone integration into treatment, for example, allowing participants to download the treatment app on their own phones or use a low-cost study smartphone (or offering both options). Long-term app engagement may be increased via built-in reminders, alerts, and in-app messages.
ClinicalTrials.gov NCT02707887; https://clinicaltrials.gov/ct2/show/study/NCT02707887
Limited access to evidence-based substance use disorder (SUD) treatment is a pervasive problem in the United States. Of the 20.3 million Americans who experienced an SUD in the past year, only approximately 12% received treatment at a specialty facility [
The Life Enhancement Treatment for Substance Use (LETS ACT) [
Despite positive initial findings, there continues to be room to improve posttreatment outcomes by increasing out-of-session treatment engagement. Although significantly lower than a contact-matched control condition, more than 50% of LETS ACT participants reported using substances by 3 months posttreatment [
Integrating smartphones into therapy is a promising strategy for increasing engagement outside group-based BA sessions. Features such as built-in guidance, prompts, and reminders can assist individuals in completing homework in a manner compliant with treatment guidelines, for example, by reminding participants to link planned activities to specific values. Users of smartphone apps for addiction recovery frequently cite the portability of apps as an advantage as well as their discreet nature [
Current research suggests that interventions involving smartphone apps are feasible and well accepted among individuals with SUDs. Research in SUD treatment samples demonstrates high rates of smartphone ownership and use, similar to the general population [
This study reports feasibility data from a trial (NCT02707887) testing the effectiveness of a smartphone-enhanced BA treatment for SUD (smartphone-enhanced LETS ACT). The aims of the study are to (1) describe the features of the app and its use and integration into treatment, (2) report participants’ self-reported feasibility and acceptability of the app, and (3) discuss challenges and provide recommendations for future smartphone app integration into behavioral treatments for SUD.
The LETS ACT app was designed to largely reflect the paper treatment materials used in previous studies of LETS ACT, with a number of added features to facilitate theory-driven treatment engagement. In the development phase, the research team drew from prior research and consultation with researchers and clinicians with expertise in SUD treatment and the development of technology to enhance behavioral interventions. The design included app features intended to address some of the limitations of paper materials (eg, providing in-app suggestions for improving homework compliance based on the user’s weekly progress). Furthermore, the app was designed to collect daily mood and substance use data. The final app was developed through an iterative piloting process, which included testing a web-based version before piloting the app with individuals in inpatient SUD treatment.
The data presented here come from a single-site, 3-arm trial conducted at an intensive outpatient SUD treatment center in Raleigh, North Carolina, comparing smartphone-enhanced BA with standard BA and treatment as usual (TAU). The focus of this analysis is to determine the feasibility and acceptability of the smartphone-enhanced treatment condition before a future report of the main outcomes of the parent trial. All participants received TAU. A total of 65 participants were randomized to smartphone-enhanced LETS ACT and attended at least one session of treatment; of these, 56 attended a second session and received a smartphone. Data for this study were collected at the pretreatment assessment, posttreatment, and at 1- and 3-month posttreatment follow-ups (FU1 and FU3). All study procedures were approved by the institutional review board.
Patients at the outpatient facility were primarily low-income individuals with a range of SUD diagnoses who voluntarily enrolled in the treatment. Patients were recruited by the research team weekly through announcements at the end of the TAU treatment groups and by approaching individuals after these groups were released. Interested individuals were assessed for eligibility, provided informed consent, and completed the pretreatment assessment. Randomization occurred at the group level using a computerized urn randomization program, and participants were blinded to the condition (ie, participants were recruited in waves and were unaware of the 3 arms of the trial). Study exclusion criteria were (1) age >65 or <18 years, (2) less than fifth grade English reading level (ie, score <42 on the Wide Range Achievement Test), (3) current impairment due to psychotic symptoms, (4) completion of >6 weeks of TAU, and (5) inability to give written informed consent to participate. Following treatment, participants completed FU assessments at the outpatient treatment facility or a public location with adequate privacy (eg, public library).
The smartphone app was developed as an adjunct to LETS ACT [
Next, emphasis shifts to an activity called Life Areas, Values, and Activities (LAVA). LAVA involves identifying activities associated with specific values and life areas (eg, education and work, emotional health, hobbies and recreation, and relationships). Participants are guided through the LAVA activity by selecting a life area that is important to them (eg, physical health), then identifying a value they hold related to that life area by answering the question, “What is important to me within this life area?” (eg, “It is important to me to increase energy and strength”). Participants then generate specific, measurable activities aligned with their values, with an emphasis on balancing enjoyable and important activities (eg, “In order to have energy and strength [value], I will walk in the park for 30 minutes [activity]”). Earlier sessions focus on tracking daily activities and creating LAVA lists. Later sessions shift focus to planning and implementing these activities in a daily plan (
Screenshots of the LETS ACT app. A: Daily Plan with completed activity; B: Life Areas, Values, and Activities; C: Activity prompt in Life Areas, Values, and Activities feature; D: Plan Ahead; E: Rating enjoyment and importance; F: Weekly Progress; G: Help page; H: Mood ratings.
Participants are given home practice assignments after each session, which include instructions for the continued use of each component. For example, after the introduction of LAVA in session 2, participants are asked to record at least one value and activity for their chosen life areas. After the introduction of the daily plan in session 3, participants are asked to plan and complete at least one activity per day for the remainder of the treatment. Participants are encouraged to continue planning and completing activities after the completion of treatment using their smartphones; however, they are not given any specific assignments to complete during the FU period.
Participants in LETS ACT-SE were provided with Apple iPhone 6 smartphones with the LETS ACT app predownloaded during the second treatment session. Phone plans were set up and paid for by the research study; plans included unlimited calls and text messages and 4 GB of wireless data per month. The intent of this service was for participants to use their phone for regular use, thus allowing the research team to assess the feasibility of the app on a personal use device. This ensured that all participants had consistent access to the LETS ACT app (which was programmed specifically for the iPhone to limit development costs) as well as to reliable internet access throughout the study duration, allowing for ongoing data collection. At this time, they were given a brief introduction to the smartphones and LETS ACT app as well as a packet of information about basic features of the phones and instructions for use (eg, how to change settings). Participants absent in session 2 were given the phone and instructions at the next treatment session attended. Participants were introduced to each app component during the sessions, with a quick therapist-led tutorial followed by in-session practice. Participants were asked to use the smartphone app to record their homework. They were informed that the smartphones were theirs to use until their FU3 appointment, at which time they returned their phones to the research team and received monetary compensation. Individuals who lost their smartphone or forgot to bring it to treatment were provided with equivalent paper forms.
All study participants were enrolled in a substance use disorder intensive outpatient program, in which treatment is based on the matrix model of intensive outpatient treatment [
Key components of the LETS ACT app include the LAVA library, Plan Ahead or Daily Plan, Weekly Progress, and Emergency button, accessible via icons on the home screen of the app (this is also the Plan Ahead screen;
The LAVA feature (
Planning value-based activities is central to the LETS ACT treatment, and the app includes 2 features that assist with this. The Plan Ahead feature (
The Daily Plan feature is the home screen of the app (
On the home screen, an option in the top right corner allows the user to view their weekly progress, that is, the percentage of planned activities completed in the previous week (
The Emergency button appears as a red siren at the top left of the Daily Plan screen (
The Help icon brings the user to a page (
On opening the app for the first time each day, the user is prompted to rate their current mood (
A questionnaire administered at pretreatment assessed smartphone ownership, use, and likelihood of using a smartphone for a research study. In addition,
Measures and outcome variables.
Construct | Variable description | Scale or possible value range | Time points |
Past-week use of app componentsa | Multiple (all past week): |
0-7 days | PTc |
Average weekly use of app components in past montha | Multiple (all past month): |
0-7 days | FU1d and FU3e |
Any app usea | Whether participant reported using any (ie, one or more) app component at least 1 day per week since the previous assessment | Yes or no | PT, FU1, FU3 |
App component usefulnessf | Degree to which participant agrees each component was a useful part of treatment: |
Scale of 1 (strongly disagree) to 5 (strongly agree) | PT |
Reasons for not using specific app componentsf | Reasons for not using: |
Can select all that apply from a list of reasons, select |
PT |
Reasons for low weekly app usef | Reasons for not using the app at least 3 times a week | Can select all that apply from a list of reasons, select |
PT |
aDescribed under Self-reported Use of App Components section.
bLAVA: Life Areas, Values, and Activities.
cPT: posttreatment.
dFU1: 1-month follow-up.
eFU2: 3-month follow-up.
fDescribed under App Component Usefulness and Reasons for Not Using section.
A questionnaire administered at posttreatment assessed participants’ self-reported app engagement during the past week. Participants indicated the number of days in the past week that they used each treatment component outside of the treatment sessions. At FU assessments (ie, FU1 and FU3), participants were given a similar questionnaire that assessed engagement with the app components during the past month. This included the average number of days per week that the participant used each component of the LETS ACT app and details about their use (eg, the number of activities scheduled and completed and the number of days per week with at least one scheduled activity;
A questionnaire administered at posttreatment assessed participant feedback about the treatment and its components (
Data were analyzed using SPSS (version 25.0, IBM Corp). First, descriptive statistics were calculated for all variables used in subsequent statistical analyses. This included means, SDs, and ranges for continuous variables and percentages for all categorical variables. To examine the feasibility and acceptability of the LETS ACT app, summary statistics (eg, mean, median, and SD) were calculated to characterize participant ratings regarding the usefulness of each app component, as well as self-reported engagement with each component (ie, past-week use of app components at posttreatment and average weekly use of app components in the past month at FU1 and FU3). Chi-square tests were used to examine differences in the proportions of participants who reported any app use at each time point. Two-tailed paired-sample
Of the 56 participants who received a smartphone, 21 (38%) were women. Overall, 61% (34/56) of participants were White and 38% (21/56) were Black. The average age was 42.4 (SD 10.5; range 24-62) years, and the participants had an average of 12.1 (SD 3.0; range 1-21) years of education. In terms of substance use, participants reported an average of 3.8 (SD 6.9; range 0-30) days of substance use in the past 30 days at pretreatment. For DSM-5 SUDs, 73% (41/56) of participants met the criteria for alcohol use disorder, 59% (33/56) met the criteria for cocaine use disorder, 45% (25/56) met the criteria for opioid use disorder, and 30% (17/56) met the criteria for cannabis use disorder. Regarding psychiatric comorbidity at pretreatment, 11% (6/56) of participants met the DSM-5 criteria for a current major depressive episode, and the average score on the Beck Depression Inventory at pretreatment was 12 (SD 11.3; range 0-51), reflecting minimal depressive symptoms. A total of 29% (16/56) of participants met the criteria for at least one DSM-5 anxiety disorder, including 18% (10/56) for social anxiety disorder, 11% (6/56) for agoraphobia and generalized anxiety disorder, 9% (5/56) for panic disorder, and 7% (4/56) for obsessive-compulsive disorder. Furthermore, 11% (6/56) met the criteria for bipolar I disorder, and 9% (5/56) met the criteria for posttraumatic stress disorder.
Of the 56 participants who received a smartphone, 54 (96%) were retained in the study through the 3-month FU assessment. One participant withdrew from the study at posttreatment because of having a busy work schedule, and a second died (of non–study-related causes) between the posttreatment and 1-month FU assessments. Overall, 27% (15/56) of participants reported at least one interruption in their ability to use their smartphone during or after treatment up to the 3-month FU, including phone lost or stolen (n=8), inability to access phones due to incarceration (n=3), and other issues (eg, technical issues with the phone; n=4).
Among participants who received a smartphone, pretreatment data from those who reported their current smartphone ownership (n=38) indicated that 79% (30/38) owned a smartphone they could use daily. In total, 14% (5/37) reported that they owned an iPhone, whereas 62% (23/37) reported having an Android phone (9/37, 24%, either provided an invalid response or reported that they did not know what type of phone they owned; the remaining participant had missing data). Overall, 76% (29/38) of participants reported that they used the internet and apps on their phones. Participants were asked how likely they would be to use a smartphone if one was provided for treatment on a scale of 1 (“I would never use it”) to 10 (“I would definitely use it”); the average rating was 8.42 (SD 2.82).
Self-reported app use data were obtained at posttreatment and FU assessments for 96% (54/56) of participants. Of these, 72% (39/54) reported any app use at posttreatment, 69% (37/54) reported any app use at FU1, and 37% (20/54) reported any app use at FU3. Chi-square tests indicated that the proportion of participants reporting any app use at posttreatment was significantly greater than the proportion reporting app use at FU3 (χ22≥20.3;
Weekly app use by component. FU: follow-up; LAVA: Life Areas, Values, and Activities; PT: posttreatment.
For each app component, participants rated their agreement with the statement that the app component was
Participants were also asked to provide open-ended feedback about the smartphone-enhanced treatment, including suggestions for improvement. Of the 51 qualitative responses, 73% (37) expressed purely positive feedback. Written comments cited the usefulness of the treatment and highlighted the novelty of the smartphone-enhanced intervention and its utility in facilitating activity planning, helping participants to “be consistent”, and supporting their recovery. Positive comments included, “It was a good way to get me thinking about how my activities affect my emotions,” “[The app] keeps me on track to be more consistent on a daily basis with responsibilities,” and “It gave me more recovery tools to work with.” Suggestions for improvement included increasing overall ease of use, “more scheduling options” for activities, and providing a smartphone for use after the study FU period.
Across all components except for the Help page, forgetting to use the app component was by far the most frequently endorsed reason for lack of use (activity scheduling: 19/56, 34%; LAVA: 10/56, 18%; Emergency: 9/56, 16%; and Weekly Progress: 10/56, 18%), whereas not having the smartphone when the participant needed to use the app was generally the second-most endorsed (activity scheduling: 5/56, 9%; LAVA: 6/56, 11%; Emergency: 5/56, 9%; and Weekly Progress: 5/56, 9%). For the Help page, the most frequently endorsed reason was lack of need for the feature (19/56, 34%), whereas forgetting and not having a smartphone were the second-most endorsed reasons (3/56 each, 5%). To provide an example of the distribution of responses for 1 main component,
Reasons for not scheduling activities (n=56).
If there were days when you did NOT have an activity scheduled, it was because: (check all that apply) | Frequency, n (%) | |
This does not apply to me because I scheduled an activity on most days. | 18 (32) | |
I did not remember to use the Daily Plan. | 19 (34) | |
I did not have the smartphone with me when I needed to fill it out. | 5 (9) | |
Filling it out took too much time or effort. | 3 (5) | |
I had technical difficulties with the smartphone. | 3 (5) | |
I did not think it would be helpful to me or my treatment goals. | 1 (2) | |
Filling it out made me uncomfortable. | 1 (2) | |
It was difficult to understand how to use it. | 2 (4) | |
|
6 (11) | |
|
Lost or do not have phone | 1 (2) |
|
Incarcerated or hospitalized | 2 (4) |
|
Other or undisclosed | 3 (5) |
Missing or no response | 13 (23) |
With regard to self-reported app use, the majority (37/54, 69%) of participants reported continuing to use their LETS ACT app until 1 month posttreatment, but this proportion decreased significantly (to 20/54, 37%) by 3 months posttreatment. Research suggests that it is typical for mobile app use to quickly drop off. Indeed, many app users use an app only once; on average, user retention is 42% after 1 month and 27% after 3 months [
Regarding specific app components, participants generally agreed that each app component was a useful part of their treatment. The LAVA library and Plan Ahead feature, which were both essential to the core homework of activity planning, were the most used components and were also rated as more useful compared with features such as the Emergency button and Help page.
In this study, when participants did not use their apps or the individual app components, they generally reported that this was because of either forgetting or not having the smartphone with them. Both reasons for nonuse may reflect the drawbacks of giving participants a study smartphone rather than downloading the app on their own phones. This highlights a critical decision in smartphone-enhanced intervention research, that is, whether to provide smartphones for each participant to ensure consistent phone access or to offer an app that participants can download on their own phones to prioritize utility, ease of use, and generalizability. Research suggests that low-income people who use drugs have high rates of smartphone ownership, but that they tend to cycle through smartphones and have inconsistent access to wireless data [
Forgetting to use the app was very common, so future studies examining the integration of smartphone apps into treatment may consider a range of strategies to mitigate this. This could include adding app features that target engagement. For example, reminders in the form of push notifications have been shown to significantly increase app use in smartphone interventions [
Participants reporting that they did not have their smartphones with them when needed may reflect challenges inherent to maintaining treatment engagement among low-income people with SUD, many of whom experience significant instability in their daily lives. Some participants were incarcerated during the FU period, whereas others had smartphones that were lost or stolen; still others reported in informal conversations with research staff that they were nervous about losing their smartphones and chose to store them in a safe place, making them inconvenient to use regularly. Future studies with this population may wish to purchase low-cost smartphones (typically Android devices), which would increase the feasibility of offering replacement devices when needed. Given that this study found that most participants owned Android devices (vs iPhones [Apple Inc]) at pretreatment, they may have the benefit of being familiar to participants in addition to being more affordable.
The results of this study must be interpreted in the context of its limitations. Reliance on self-reported measures of app engagement may be associated with recall bias. In addition, the study sample was recruited from an intensive outpatient treatment center serving a primarily low-income, high school–educated clientele, and participation in the current intervention was offered in addition to standard treatment; the results may not be generalizable to other populations or treatment settings.
There is a clear need for evidence-based SUD treatments that can be delivered at a low cost, and it is essential to find new and effective ways to engage participants in these treatments. Despite notable growth in the area of app-based psychological interventions, very little research has examined the impact of introducing smartphones into in-person therapy, especially in the area of SUD treatment. This study found evidence that integrating a smartphone app into a BA treatment for SUD is feasible and well accepted and that participants found the app useful for planning value-based activities, which is the core task of BA. However, the study also found that engagement with the app decreased over the FU period and that participants frequently reported forgetting to use the app, highlighting the need for further efforts to sustain out-of-session engagement over time. These findings must be interpreted in light of the specific study methods (eg, the provision of study smartphones to all participants); future research is needed to examine differences in app use when allowing participants to download treatment apps on their own smartphones. Studies are also needed to examine specific treatment contexts and participant characteristics that may be associated with receiving more benefit from smartphone-enhanced interventions.
behavioral activation
follow-up
Life Areas, Values, and Activities
Life Enhancement Treatment for Substance Use
substance use disorder
treatment as usual
None declared.