Effect of Mindfulness-Based Cognitive Therapy Combined with Lifestyle Modification versus Lifestyle Modification Alone on Ovarian Sonographic Morphology in Reproductive-Age Women with Polycystic Ovary Syndrome: A 12-Week Randomized Clinical Trial

Does adding Mindfulness-Based Cognitive Therapy (MBCT) to a 12-week lifestyle modification program, compared with lifestyle modification alone, result in greater improvement in ovarian morphology (based on changes in follicle number, ovarian volume, and other structural characteristics) in reproductive-age women with Polycystic Ovary Syndrome (PCOS)?

Participants are randomly assigned (using block randomization) to two groups. The study is single-blind. The sample size is 52 individuals per group.

The interventions will be delivered online through educational and counseling sessions. Participants will be recruited from Shahid Motahari Subspecialty Clinic in Shiraz, and ultrasound assessments will be performed at Tabesh Medical Imaging Center in Shiraz. The radiologist performing the ultrasound examinations, image analysts (manual and artificial intelligence-based), and statistical analyst will be blinded to group allocation.

Inclusion criteria: Women aged 18–35 diagnosed with PCOS based on Rotterdam criteria. Exclusion criteria: Use of hormonal medications in the last three months, pregnancy or breastfeeding, and severe psychiatric disorders or other uncontrolled endocrine disorders.

Lifestyle Modification Group: Involves nutrition education, physical activity, behavior change strategies, and regular participant follow-up, conducted over 12 weekly sessions. Counseling Group: In addition to receiving lifestyle modification training, participants receive 12 sessions of MBCT counseling.

Ovarian follicle number; ovarian volume; perceived stress; anxiety; quality of life related to Polycystic Ovary Syndrome

Participants will be allocated to the two study groups using stratified block randomization with a 1:1 allocation ratio. The unit of randomization will be the individual participant. To ensure balance in participant allocation according to body mass index (BMI), randomization will be stratified into two BMI strata: BMI <30 and BMI ≥30. For each stratum, an independent random allocation sequence will be generated by an independent statistician using R statistical software and the blockrand package. Within each stratum, variable block sizes of 4 and 6 will be used, and block sizes will be randomly determined. Each block will contain an equal number of allocations to the two study groups, and the order of allocations within each block will be randomly determined by the software. After eligibility has been confirmed and the participant has formally entered the study, the participant will be assigned, according to the relevant BMI stratum, to the next available position in the allocation list for that stratum. Group assignment will then be determined according to the pre-generated random sequence. The random allocation sequence will be generated by an independent statistician, and the researcher responsible for participant recruitment will not have access to the allocation sequence during the baseline assessment. To ensure allocation concealment, the random allocation sequences for each BMI stratum will be placed in opaque, sealed, light-proof envelopes with sequential numbering. The envelopes will be prepared separately for each BMI stratum according to the allocation order. The envelope corresponding to each participant will be opened only after final confirmation of eligibility criteria, obtaining informed consent, and formal entry into the study. Therefore, group allocation cannot be predicted or altered before the participant formally enters the study.

The statistician, radiologist (ultrasound assessor), and imaging analyst performing the AI-based ultrasound image analysis will be blinded to participants’ group allocation. Participants and the intervention provider will not be blinded due to the nature of the intervention. To maintain blinding: • DICOM files are coded and stored without reference to the intervention group prior to analysis. • The statistical dataset is analyzed using group codes (A/B) without disclosing the nature of the intervention.

There is no further information