The end of problem gambling, Only a push away?

The primary objective is the exploration of the added benefits of Approach Bias Modification (AppBM) atop an online CBT program in reducing or stopping gambling immediately after the intervention and several (three and six) months after. To this end, we designed the present study to contain two conditions. Participants in both conditions will receive 9 sessions of online CBT. Participants in the experimental condition also receive 9 APPBM sessions, atop the CBT, whilst the participants in the control condition receive 9 AppBM-placebo sessions. Both CBT and APPBM sessions are administered online.



We expect participants in both conditions to show a reduction in gambling problem severity. However, we expect the reduction to be greater amongst participants in the experimental (appBM) condition. We also expect the approach bias to decrease or reverse in the experimental condition and this change to mediate the greater decrease in gambling problem severity in the experimental condition.



We will also address the moderating effects of trait impulsivity and the strength of the approach bias before treatment, on the relation between AppBM and gambling problem severity. Based on previous research, we expect participants with strong a strong approach bias and/or high impulsivity to benefit more from AppBM than do participants with a weaker approach bias and/or low impulsivity.

Hypothesis 1: Participants in both conditions (experimental and placebo) will show a reduction in gamble problem severity immediately post-intervention and at 3-and 6-month follow-ups. However, participants in the experimental condition will show a greater reduction than do participants in the control condition.



Hypothesis 2: Participants in the experimental condition will show a decreased approach bias immediately post-intervention and at 3 month follow-up. Participants in the control condition are expected to show no decrease in approach bias.

Hypothesis 3 (moderated mediation): Change in approach bias will mediate the change in gambling severity problems. Participants with a stronger approach bias and/or higher impulsivity at baseline will benefit more from APPBM than participants with a weaker approach bias and/or lower impulsivity at baseline.

T0: registration (0)

T1: mid-assessment (T0 + 5 weeks)

T2: post-assessment (T0 + 9-12 weeks)

T3: follow-up I (T2 + 12 weeks)

T4: follow-up II (T2 + 24 weeks)

1. Online Cognitive Behavioral Therapy

Participants in both conditions receive 9 sessions of online CBT. The CBT protocol is an adaptation of the Dutch CBT protocol for the treatment of substance abuse and gambling. Some minor adjustments to the protocol have been made to better meet the needs of online treatment of gambling disorder. In the sessions, the following topics are addressed (in this order): (1) Introduction & inventory; (2) Self-control and inciters; (3) Function analysis; (4) Assessment of risk-situations and plan of action; (5) Emergency-plan and self-reward; (6) Coping with urge/relapse; (7) Coping with and changing of cognitions; (8) Irrational beliefs & concepts of chance, coincidence and redistribution rate. Each sessions follows a similar pattern: (1) introduction and evaluation of last week (using data from the online diary, CBT assignments, APPBM feedback and from participant); (2) sessions main topic; (3) discussion of next week, homework assignments and closing of the session. Sessions are planned with the participant and take place online, trough Gokhulp, as a chat between participant and therapist. Participants are asked to keep an online diary of their gambling activities. This diary is used by the therapist to monitor gambling behavior and to increase participants’ insight. If after 9 sessions the therapist feels that the participant is in need of some more treatment, a maximum of 3 extra sessions can be added addressing one or more of the following topics: (1) Social skills; (2) Coping with somber emotions; (3) Coping with stress; (4) Problem solving. Therapists are trained professionals in administering CBT, are experienced in this medium of treatment and follow chat-etiquette described in the protocol.



Records of the chat sessions are stored (anonymous) and a portion (10%) will be independently coded for treatment adherence by two coders. Coders will score whether or not session-specific topics, as described in the treatment manual, are addressed (yes/no) and will score the therapists’ therapeutic attitudes and style on a 7 point scale (1 being “very low” and 7 being “very high”). Coder reliability will be assessed.

2.1 Approach Bias Modification
Automatic approach tendencies towards gambling are assessed and trained with the modified Approach-Avoidance Task. The AAT is a computerized speeded reaction-time task in which participants are asked to react to the stimulus presentation format and ignore the stimulus content (i.e., irrelevant-feature implicit measure).

In this task, a gambling-related picture or a control picture is presented in the centre of the screen. The picture is three degrees tilted to the left or to the right. Participants are instructed to respond to the tilt direction of the picture by pushing pictures tilted to the left away from them and pulling pictures tilted to the right towards them. The combination of picture tilt direction and response (left/push and right/pull versus left/pull and right/push) is counterbalanced across participants. Participants’ response is accompanied by a zooming effect, which increases the picture size in the pulling closer response and decreases it in the pushing away response, creating the perception of approaching or pushing away (avoiding) the stimulus. Stimulus stays on screen for 3000ms; in case of no response the trial is restarted after repeating the task instructions.
In the AAT assessment version and in the assessment and placebo block of the retraining, gambling and controls pictures are presented equally often in both formats. An approach bias score is computed by subtracting the difference in median reaction time between pull responses and push responses to control stimuli from the difference in median reaction time between pull responses and push responses to gambling-related stimuli. The score provides a relative measure of gambling approach bias compared to control cues. Positive scores indicate an approach tendency, negative ones an avoidance tendency. In the Approach Bias Modification block, participants are trained to avoid gambling cues by exposing them only to gambling/push and non-gambling/pull trials.

Stimuli are pairs of matched gambling and non-gambling pictures, which are counterbalanced for presentation format only in the assessment stage. The stimuli stay on screen for 3000ms; in case of no response the trial is restarted after repeating the instruction.

2.2Placebo Retraining
The control group will complete a placebo retraining consisting of a continuous assessment task by randomly receiving a longer version of the approach bias assessment task with a constrained 50/50 proportion. Gambling and control pictures are presented in all trial formats.