Sleep predicts the response to rTMS and CBT in patients with OCD: an open label effectiveness study

Priya T Gajadien, Tjardo S Postma, Iris van Oostrom, Karel W F Scheepstra, Hanneke van Dijk, Alexander T Sack, Odile A van den Heuvel, Martijn Arns

Research output: Contribution to journal/periodicalArticleScientificpeer-review

3 Citations (Scopus)
71 Downloads (Pure)


BACKGROUND: Although many OCD patients benefit from repetitive transcranial magnetic stimulation (rTMS) as treatment, there is still a large group failing to achieve satisfactory response. Sleep problems have been considered transdiagnostic risk factors for psychiatric disorders, and prior work has shown comorbid sleep problems in OCD to be associated with non-response to rTMS in OCD. We therefore set out to investigate the utility of sleep problems in predicting response to rTMS in treatment resistant OCD.

METHOD: A sample of 61 patients (treated with 1-Hz SMA or sequential 1-Hz SMA+DLPFC rTMS, combined with cognitive behavioral therapy) were included. Sleep disturbances were measured using the PSQI, HSDQ and actigraphy. Treatment response was defined as a decrease of at least 35% in symptom severity as measured with the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS).

RESULTS: 32 of 61 patients (52.5%) responded to rTMS, and trajectories of response were similar for both rTMS protocols. Three PSQI items (Subjective Sleep Quality; Sleep Latency and Daytime Dysfunction) and the HSDQ-insomnia scale were found to predict TMS response. A discriminant model yielded a significant model, with an area under the curve of 0.813.

CONCLUSION: Future replication of these predictors could aid in a more personalized treatment for OCD.

Original languageEnglish
Pages (from-to)100353
JournalInternational journal of clinical and health psychology : IJCHP
Issue number2
Publication statusPublished - 2023


Dive into the research topics of 'Sleep predicts the response to rTMS and CBT in patients with OCD: an open label effectiveness study'. Together they form a unique fingerprint.

Cite this