Digital relapse prevention plan for substance use disorders: study protocol for a multicentre randomised controlled trial

Abstract

Introduction

Relapse remains a major challenge in the treatment of substance use disorders (SUDs), particularly during follow-up. Digital tools are emerging as supportive resources, but few deliver real-time interventions. This study will examine the effectiveness of a digital relapse prevention plan (DRPP) integrated into a certified mobile application to detect early risk signs and provide immediate, personalised responses.

Methods and analysis

A multicentre randomised controlled trial will recruit adults with SUD. Participants will be randomised to standard treatment plus a restricted app version (control) or the same treatment with the full app, including automated alerts and DRPP access (experimental). The plan can be activated manually or automatically through smartphone sensors detecting risk patterns. The primary outcome will be time to first clinical relapse, while secondary outcomes will include patient satisfaction with the DRPP, adherence and perceived emotional self-regulation. Findings are expected to provide robust evidence on the feasibility, acceptability and clinical utility of digital relapse prevention strategies.

Ethics and dissemination

This study obtained ethical approval (code 25/327) from Committee of Hospital Universitario 12 de Octubre.

Trial registration number:

NCT07052175.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Relapse remains a major barrier in substance use disorder treatment, and conventional follow-up methods often fail to provide real-time, personalised support. Mobile health tools have shown potential, but few have been tested through rigorous randomised controlled trials in this field.

WHAT THIS STUDY ADDS

• This trial examines, for the first time in Spain, the effectiveness of a digital relapse prevention plan (DRPP) integrated into a certified mobile application, combining continuous monitoring, automated alerts and tailored interventions. It also explores patient satisfaction, adherence and the potential of digital phenotyping in addiction care.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• If effective, the DRPP could be integrated into routine multidisciplinary care as a scalable, patient-centred strategy to improve long-term recovery and reduce healthcare burden.

Introduction

Substance use disorder (SUD) remains one of the major challenges in public health. The biopsychosocial model currently represents the gold standard of care, integrating medical detoxification, psychotherapy, occupational therapy and social support.^{1 2} Interdisciplinary collaboration has been shown to improve adherence and outcomes.³ Among psychotherapeutic approaches, cognitive behavioural therapy, motivational interviewing and relapse prevention therapy stand out as effective strategies to modify dysfunctional thinking patterns, strengthen motivation and manage high-risk situations.⁴,⁷

Despite these advances, relapses remain a common and serious challenge. Up to 50% of individuals undergoing treatment experience relapse,⁸ underscoring the need for enhanced post-treatment monitoring. The consequences of relapse can be severe, including job loss, family breakdown and return to harmful patterns of substance use.⁹ For this reason, implementing continuous and individualised follow-up strategies may be critical. In this line, early detection of relapse warning signs allows for timely intervention.^{10 11}

However, conventional follow-up methods often fall short in real-time responsiveness and personalisation. Emerging digital health technologies, particularly mobile applications, offer promising solutions in this context. These tools enable repeated, live monitoring data collection on users’ behaviours and experiences, minimising recall bias and maximising ecological validity.^{12 13} Immediate tracking, mobile-delivered interventions during high-risk moments have been shown to be effective, providing a personalised and immediate response that enhances clinical care.¹⁴

Moreover, such tools have demonstrated value in research on alcohol and tobacco use, capturing behavioural patterns based on real-time data (such as mood fluctuations, frequency of use, geolocation, self-reported cravings), which cannot be detectable through retrospective questionnaires.¹⁵,¹⁷ In this sense, Bayesian predictive models derived from these behavioural patterns have shown success in identifying key relapse risk factors.¹⁷,¹⁹

These innovations enhance and complement traditional addiction treatment frameworks. In a highly complex clinical and social landscape, the integration of biopsychosocial approaches with a digital tool offers a unique opportunity to redefine care paradigms in mental health and addiction. This integration places the patient at the centre of the therapeutic process, promoting empowerment, autonomy and improved quality of life.

Aim and objectives

This study aims to examine the clinical effectiveness of a digital relapse prevention plan (DRPP) integrated into a mobile application as an adjunct to multidisciplinary treatment for SUD. The objectives are to compare clinical outcomes between DRPP users and controls; determine the effectiveness of automated risk detection and DRPP activation; examine patient satisfaction with the app and DRPP (usefulness, usability, self-care, self-experience); and identify adherence patterns and their association with treatment continuity and outcomes.

Methodology

Study design

This study is a prospective, multicentre, randomised controlled trial involving adult individuals diagnosed with SUD, with or without polysubstance use. Participants will be randomly assigned (1:1) using a centralised web-based system with concealed allocation to ensure unbiased group assignment. A total of 120 participants will be recruited from specialised addiction treatment centres within the San Juan de Dios network in Spain. Eligible participants will be adults diagnosed with disorders classified under the International Classification of Diseases, 10th revision category of mental and behavioural disorders due to psychoactive substance use.²⁰ Inclusion criteria: (1) adult between 18 and 65 years of age; (2) understanding Spanish; and (3) initiating treatment in addiction units. Exclusion criteria: have severe medical conditions or are unable to sign the informed consent form.

The target sample size of 120 participants was determined to provide approximately 80% power (α=0.05) to detect a clinically meaningful HR of about 0.65 in time-to-event analyses, assuming a 50% relapse rate at 12 months and up to 20% attrition.

Materials

Outcome variables will include withdrawal symptoms, craving, emotional well-being, psychosocial functioning, relapse rates and patient satisfaction. Feasibility, acceptability and clinical utility of the DRPP in follow-up will also be assessed (table 1).

Table 1. Outcome measures that will be collected and analysed.

Outcome	Measure	Description	Time frame
Descriptive measures	Sociodemographic data	To characterise the participants’ sociodemographic profile, data will be collected on key variables such as age, gender, marital status, level of education and current occupation. This information will be gathered during the initial assessment phase, allowing for a comprehensive description of the sample’s social and demographic context.	At the beginning of the intervention (baseline assessment)
Primary outcome measure	Self-reported relapse during follow-up via telephone	To assess the occurrence of substance use relapse after the transition to the follow-up phase (when the participant is no longer in the clinic), a structured telephone interview will be conducted. Participants will be asked whether they have experienced any relapse since entering the follow-up phase (Yes/No), and if so, when the relapse occurred. This method allows monitoring of relapse events that may not be detected through biological testing during this period.	During the follow-up phase (telephone interviews at 6 and 12 months after baseline)
	Self-reported relapse during follow-up via app notification	Brief and structured self-administered assessment delivered through the mobile application, designed to monitor relapse events. This assessment is triggered 1 week after a relapse alarm is detected or the DRPP is activated, with the aim of assessing whether a relapse has occurred. Participants will receive a notification prompting them to answer the following question: ‘We have detected a possible alert, and we would like to confirm whether you have experienced any risk situation during the past week. Has this been the case?’ (Yes/No). If the response is affirmative, the app will display a follow-up question: ‘Did this situation lead you to use substances?’ (Yes/No).	Throughout the intervention and follow-up phases (baseline to 12 months)
Secondary outcome measures	Quality of Life (WHOQOL-BREF)	A 26-item instrument assessing perceived quality of life across four domains: physical health, psychological well-being, social relationships and environmental context. Items are rated on a 5-point Likert scale, and domain scores are transformed to a 0–100 scale. Demonstrates good internal consistency, with Cronbach’s alpha=0.90.	Baseline, post-treatment (3 months) and follow-up (6 and 12 months)
	Purpose in Life Test—Short Form (PIL-10)	A 10-item self-report scale assessing life purpose and personal meaning. Responses are rated on a 7-point Likert scale, with total scores ranging from 10 to 70. Higher scores reflect greater perceived purpose. The scale includes two dimensions: (1) life satisfaction and meaning, and (2) goals and direction. Shows good psychometric properties (α=0.85 overall; 0.84 for life satisfaction and 0.69 for goals and purpose).	Baseline, post-treatment (3 months) and follow-up (6 and 12 months)
	Patient Health Questionnaire-9 (PHQ-9)	A 9-item self-report tool assessing depressive symptom severity over the past 2 weeks, based on DSM-IV criteria. Each item is scored from 0 (not at all) to 3 (nearly every day), yielding a total score of 0–27. A cut-off score of 10 yields sensitivity of 0.78 and specificity of 0.87. Spanish versions show excellent reliability (α=0.78–0.90), with current data showing α=0.93.	Baseline, post-treatment (3 months) and follow-up (6 and 12 months)
	Beliefs About Substance Use and Craving Beliefs Questionnaire	25 items self-report questionnaire designed to assess maladaptive beliefs related to substance use and beliefs about substance craving. Each item is rated on a 5-point Likert scale ranging from ‘totally disagree’ to ‘totally agree’. Higher total scores reflect stronger dysfunctional beliefs associated with substance use (eg, perceived benefits or inevitability of use). The instrument shows excellent internal consistency (Cronbach’s alpha=0.87).	Baseline, post-treatment (3 months) and follow-up (6 and 12 months)
Other prespecified outcomes	Patient satisfaction with the digital application	27-item ad hoc questionnaire using a 5-point Likert scale (scale from 1 to 5, with 1 being ‘Not at all’ and 5 being ‘Very much’) to assess patient satisfaction with the application's usability, safety and overall experience. Higher scores indicate greater satisfaction.	Post-treatment (3 months)

DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.

Procedure

Eligible patients will be identified by the clinical team during the initial phase of their treatment. After receiving detailed information about the study, individuals who meet inclusion criteria and provide written informed consent will be enrolled and randomly assigned to either the experimental or control group using a computer-generated randomisation sequence. Participation will be voluntary.

On enrolment, each participant will undergo a comprehensive baseline assessment that includes standardised clinical, psychological and sociodemographic measures. All participants will follow a standard multidisciplinary treatment pathway based on a biopsychosocial model, consisting of medical detoxification, psychological therapy, social support and rehabilitation interventions.

Participants in both groups will be provided with the eB2 MindCare mobile application. However, only those in the experimental group will receive the fully functional version of the app, which includes real-time monitoring, automated risk alerts and access to the DRPP (table 2). In contrast, the control group will receive a restricted version of the app, with basic monitoring capabilities but no alert system or access to the DRPP. The application will operate continuously in the background, collecting both passive (eg, mobility, sleep, phone use) and active (eg, self-reported mood and stress) data.

Table 2. Content of the digital relapse prevention plan.

Digital relapse prevention plan
	Relaxation	Section dedicated to emotional and sensory regulation. Includes techniques and content such as: Guided breathing exercises. Mindfulness practices. Progressive muscle relaxation techniques. Five senses awareness exercise. Relaxing images. Relaxing videos. Therapeutic and relaxing music.
	Personal reminder	A space where the patient can record strategies to cope with difficult moments. Each situation can be personalised with: Title and description. Personal goals. Barriers and possible solutions. Complementary multimedia content: personal audio recordings, notes, images, videos, suggested activities. Examples of situations: ‘When I feel sad’, ‘When I have craving’, ‘When I feel lonely’.
	Information and psychoeducation	Educational material structured by mental health professionals to reinforce knowledge and understanding of emotions and symptoms. Includes: Infographics. Explanatory videos. Informative presentations (PDF/PPT). Short texts on emotional regulation, self-care and addictions.
	Emotional journal	A space for spontaneous emotional expression. The patient can record: Main emotion or current physical sensation. Brief description of the context or situation. Reflection: What can I do now? Option to record personal audio for emotional release. Other free-form entries such as drawings, lists or random ideas.
	Button to call support person	A direct access button to contact a previously registered family member, friend or emotional support person.
	Button to call helpline/emergency services	A direct access button to make an immediate call to a clinic, specialised centre or emergency services (based on the patient’s previous configuration).

When the system identifies marked deviations in behavioural or emotional functioning, such as reduced physical activity, disrupted sleep or persistent distress, it generates automated alerts that trigger activation of the DRPP in the experimental group.

To ensure reliable risk detection, the application requires an initial baseline monitoring period of approximately 3 weeks. During this phase, the algorithm continuously collects passive and active data to establish each participant’s individual behavioural and emotional patterns. Deviations from these personalised baselines are then used as thresholds to trigger automated alerts. This adaptive calibration minimises false positives and false negatives and enhances the clinical validity of relapse risk detection. The algorithm underlying this process has been previously developed and is currently under review for publication.

The indicators monitored include atypical activity levels (either excessive or markedly diminished), instability in mobility and sleep-wake cycles, unusually prolonged time spent at home or outside, signs of emotional dysregulation, exposure to risk situations or adverse events, abrupt changes in smartphone use (excessive or erratic) and information provided through structured questionnaires or personal notes that reveal relevant alterations in the patient’s mental or behavioural state (figure 1).

Figure 1. Digital phenotyping, behavioural profiling and change-point detection process used for relapse prediction and validation. FN, false negative; FP, false positive; TP, true positive.

Patients may also activate the plan manually at any time they perceive a need for additional support. The DRPP provides a structured set of tools for emotional self-regulation and relapse prevention, combining both automatic and patient-initiated activation pathways (figure 2).

Figure 2. DRPP. Workflow of the app-based relapse prevention intervention showing automatic and self-reported alerts, warning messages and activation of the digital plan. DRPP, digital relapse prevention plan.

Follow-up assessments will be conducted at 3, 6 and 12 months using standardised questionnaires. Additionally, structured telephone interviews will be carried out at 6 and 12 months to confirm relapse and treatment continuity, thereby complementing the standard follow-up measures and retention. Relapse events will be validated through a multimodal approach combining self-report, clinical judgement during structured interviews and, when available as part of the therapeutic process or routine follow-up, objective biomarkers such as urine toxicology or breathalyser testing.

All data collected via the app will be pseudonymised and securely stored on encrypted servers, ensuring participant confidentiality and compliance with data protection regulations. At the conclusion of each participant’s study period, app data collection will automatically cease, and participants will receive a final notification informing them that the monitoring has ended and that they may uninstall the app if desired. The database is hosted on secure servers located within the European Union, under EU data protection law. All app data will be permanently deleted once the analyses are completed.

Statistical analysis

The primary outcome is time to first clinical relapse, defined as the first episode initially detected by the app, clinically verified and, during follow-up, confirmed by structured telephone interview. Analysis will be conducted using a Cox proportional hazards model including random effects to account for individual heterogeneity and adjusted for clinical and demographic covariates. An interaction between treatment group and time will be included to explore differential effects during follow-up.

Secondary outcomes, such as patient satisfaction, perception of self-care and emotional well-being, will be analysed using linear mixed-effects models, incorporating fixed effects for group, time and their interaction, as well as relevant covariates.

Usage behaviour and adherence will be examined with latent class analysis to identify subgroups based on app interactions and DRPP activations. Multilevel mediation models will then test whether adherence mediates the relationship between intervention exposure and outcomes (relapse, quality of life, perceived benefit).

All analyses will follow the intention-to-treat principle. Missing data will be addressed with multiple imputation and sensitivity analyses. Effect sizes with 95% CIs will be reported, using validated statistical software (eg, R or Stata), with significance set at p<0.05.

Discussion

This study will examine the clinical effectiveness, feasibility and patient experience of a DRPP integrated into a mobile application for individuals with SUDs. By combining continuous monitoring of behavioural and emotional states with tailored interventions, the DRPP seeks to address the persistent challenge of relapse prevention. It is expected to extend the time to first relapse, improve emotional regulation and self-efficacy and enhance adherence by offering immediate access to coping strategies and motivational resources, thereby bridging gaps between clinical appointments.

In addition to preventing relapse, the DRPP may improve patient-reported outcomes such as satisfaction, perceived benefit and quality of life, supported by customisable features like reminders, diaries and support contacts. The study will also contribute to digital phenotyping in SUD, analysing behavioural data to identify adherence patterns and subgroups and exploring how engagement levels influence outcomes. Ultimately, the DRPP appears feasible and scalable, integrating smoothly into existing workflows while reducing clinician workload.