Abstract
Repetitive peripheral magnetic stimulation (rPMS) has been used in dysphagia rehabilitation, but comparative evidence and treatment heterogeneity remain unclear. We conducted a single-center retrospective propensity score-matched cohort study of inpatients undergoing dysphagia rehabilitation. Patients treated with rPMS plus individualized usual rehabilitation were matched 1:1 to historical controls treated before rPMS implementation (50 vs. 50). The primary outcome was change in Food Intake LEVEL Scale (FILS) from admission to discharge. Age-related heterogeneity was examined using a treatment-by-age interaction and exploratory stratified analyses (< 80 vs. ≥ 80 years). One control patient lacked discharge FILS and was excluded from the primary analysis (n = 99). Overall, rPMS was not associated with greater FILS improvement (+ 0.48, 95% confidence interval - 0.26 to 1.22; p = 0.202). Exploratory interaction analysis suggested age-related heterogeneity (interaction p = 0.008): in stratified analyses, rPMS was associated with greater improvement among patients aged < 80 years (+ 1.70, 95% confidence interval 0.34 to 3.07; p = 0.015), whereas no association was observed among patients aged ≥ 80 years (-0.52, 95% confidence interval - 1.44 to 0.39; p = 0.258). Cognitive and total Functional Independence Measure change scores were higher in the rPMS group, but residual confounding remained plausible. In this matched observational study, rPMS was not associated with overall improvement in functional oral intake measured by FILS. Exploratory analyses generated a hypothesis of possible age-related heterogeneity, but this signal should not justify age-based treatment selection.