Web Based Cognitive Bias Modification for Alcohol Addiction - From controlled trial to real-world application
Melissa Laurens is a PhD student in the Department of Psychology, Health & Technology. (Co)Promotors are prof.dr. E.T. Bohlmeijer from the Faculty of Engineering Technology (UT), dr. M.E. Pieterse from the Faculty of Behavioural, Management and Social Sciences (UT), prof.dr. E. Salemink, Utrecht University and prof.dr. S. Ben Allouch, University of Amsterdam.
Alcohol Use Disorder (AUD) is a highly prevalent and debilitating condition with severe health, social and economic consequences. While evidence-based interventions such as Cognitive Behavioural Therapy (CBT) and Motivational Interviewing (MI) are effective outpatient care options that focus on enhancing cognitive control and addressing ambivalence towards alcohol use, they predominantly target conscious cognitive processes. Dual-process models highlight the role of both impulsive and reflective systems in addiction, with approach bias (an automatic tendency to move towards alcohol-related stimuli) being an important factor thought to contribute to maintaining AUD. Approach Bias modification (ApBM) has shown promise in reducing relapse rates in inpatient settings, but its effectiveness in outpatient care, where over 80% of Dutch AUD treatment takes place, remains unclear. Furthermore, most ApBM interventions have relied on computer-based delivery, which limits accessibility.
This thesis therefore aimed (1) to evaluate an online ApBM intervention as an adjunct to outpatient treatment as usual for AUD, and (2) to develop and test a mobile ApBM application to increase accessibility.
Part 1 – Online ApBM in outpatient treatment
Chapter 2 – Protocol for a randomized controlled trial
This chapter outlines the design of a double-blind randomised controlled trial evaluating the addition of online ApBM to TAU for outpatients with AUD. The study compared eight sessions of active ApBM with placebo training over five weeks, assessing alcohol consumption, approach bias and adherence.
Chapter 3 - RCT results
A total of 139 participants were randomised. Both ApBM and placebo groups significantly reduce alcohol consumption over time, achieving safer drinking levels and lower weekly intake, indicating TAU alone is effective. The addition of ApBM did not yield significant added benefit in drinking outcomes. However, ApBM participants showed a significantly greater reduction in alcohol approach bias (d-scores) than placebo partipants, although this effect was nog consistently robust across all multiple-imputation analyses. Adherence was high overall (65% completed ≥ 6 sessions), and hights among those receiving TAU online (85.5% versus 51.2% in face-to-fae TAU). These findings suggest that while online ApBM can reduce approach bias in outpatients, this does not translate into clear incremental benefits for alcohol consumption when combined with TAU.
Part 2 – Development and testing of the mobile ApBM app
Breindebaas
Chapter 4 – Feasibility in problem drinkers
In a self-selected sample of 410 problem drinkers, three weeks of mobile ApBM use led to an average reduction of 7.8 standard units of alcohol per week, with an additional 6.2-unit reduction at three-month follow-up. These sustained changes suggest potential for supporting behaviour change outside formal treatment. However, without a control group, causality cannot be established.
Chapter 5 – User-centered development
Development of Breindebaas followed user-centered design principles, incorporating feedback from (ex-)AUD patients, problem drinkers, and addiction professionals. Key features included a customisable image library, clear explanations of the ApBM mechanism, adjustable reminder settings, and motivational feedback. These elements aimed to improve engagement and personal relevance. While users expressed interest in gamification, such features were excluded at this point to maintain therapeutic focus an avoid resource constraints.
Chapter 6 – Feaibility for MID-BIF populations
A personalised version of Breindebaas was tested with individuals with MID-BIF and AUD. The measurement function showed moderate test-retest reliability and stable reaction times. Participants found the app engaging and easy to use, with potential benefits for increasing awareness of drinking behaviour. Concerns included the possibility of craving triggers and specific usability barriers, underscoring the need for further adaptation before broader implementation.
Chapter 7 – General discussion
Online ApBM as an add-on to TAU reduced approach bias but did not improve drinking outcomes beyond TAU in outpatient care. The mobile app showed promise for reducing alcohol consumption in non-treatment populations and was well-received by a broad population. Key strengths include the rigorous RCT design, ecological validity in real-world outpatient settings, and the integration of user feedback into mobile app development. Limitations include reliance on self-reported alcohol use and absence of control groups in mobile studies. Implications of practice highlight that ApBM may be most effective as a flexible, scalable tool to complement existing treatments, especially when integrated into personalised, mobile-based interventions. Future research should focus on optimising delivery formats, exploring mechanisms of bias change, and identifying populations and contexts where ApBM offers the greatest clinical benefit.
