Functional magnetic stimulation (FMS) of the thoracic nerve roots to simulate cough has been suggested as a treatment approach in patients unable to voluntarily activate the abdominal muscles. However, factors that could influence the efficacy of FMS in clinical use have not been evaluated. In the present investigation we studied train length, posture, and frequency to determine the optimal stimulation protocol. We also evaluated the use of a valve at the mouth to enhance glottic function and investigated whether lung volume at the time of stimulation would influence the tension generated by the abdominal muscles. Studies were performed using a Magstim rapid stimulator augmented by four booster packs in nine healthy subjects; we measured the change in gastric (DeltaPga(FMS)), esophageal (DeltaPes(FMS)), and mouth pressure and expiratory flow. With our apparatus pressure generation was maximized by having a train length of at least 300 ms and a frequency of 25 Hz. Posture and valve use were not important determinants of DeltaPga(FMS) or DeltaPes(FMS). Lung volume exerted only a minor influence on DeltaPga(FMS), but the ratio DeltaPes(FMS):DeltaPga(FMS) was increased at TLC compared with FRC. Expiratory flow was increased by adopting a seated posture and using an occlusion valve with an opening threshold close to the maximum DeltaPes(FMS) generated by the stimulus train; however, expiratory flow was susceptible to interference from glottic incoordination. Representative results (with train length 600 ms, 25 Hz, and 100% power, seated) were mean DeltaPga(FMS), 166 cm H(2)O; mean DeltaPes(FMS), 108 cm H(2)O; and mean expiratory flow, 311 L/min. We confirm that FMS of the abdominal muscles can generate a substantial positive intra-abdominal and intrathoracic pressure and, consequently, expiratory flow in normal subjects.