Efficacy of deep transcranial magnetic stimulation with H7 coil in gambling disorder: COMET-7 study protocol

Methods

Single-centre, prospective, double-blind, parallel-group, randomised, sham-controlled trial with concealed allocation and blinded outcome assessment at the University Psychiatric Clinic Sveti Ivan. Men aged 18–64 years with ICD-10 F63.0 GD, right-handed, entering treatment are eligible. Key exclusions include psychosis, recent substance use disorders (except nicotine) and major neurological disease. Participants will be randomised 1:1 by minimisation (Pocock–Simon) on prespecified factors. The active arm receives FDA-pattern H7-coil dTMS: 20 Hz, 2-s trains, 20-s inter-train, 2,000 pulses/session at 100 % motor threshold, five sessions/week for five weeks plus four sessions in week six (total 58,000 pulses); the sham arm receives device-matched sham. All patients continue standard multidisciplinary care, held constant across arms. Outcomes are assessed at baseline and six weeks; selected measures are monitored weekly. The primary endpoint is change in Gambling Symptom Assessment Scale (G-SAS). Secondary endpoints include craving (GUS, VAS), PG-YBOCS, PGSI, SOGS, abstinence, depressive and anxiety symptoms (BDI-II, BAI), quality of life (QOLI), HRV parameters, and adverse events. The intention-to-treat analysis tests the group×time interaction using mixed-model ANCOVA with prespecified covariates. Sensitivity analyses include per-protocol and robust models. Missing outcomes in ITT are handled under a conservative no-effect assumption; covariate missingness is multiply imputed. Target sample size is 60 (30 per arm) to detect a moderate effect with 80 % power, allowing 15 % attrition. The trial will be registered on ISRCTN; ethics approval will be obtained before enrolment