BACKGROUND Shoulder subluxation is a frequent complication after stroke causing joint instability, shoulder pain, decreased activities of daily living, and impedance to rehabilitation progress. Electrical stimulation (ES) is considered an effective modality to reduce shoulder subluxation in acute stroke. However, few studies have investigated the effect of position-triggered ES, which induces active muscle contraction though accurate motion detection. OBJECTIVE The aim of this study was to investigate whether position-triggered ES was more effective in reducing acute hemiplegic shoulder subluxation after stroke than passive ES. METHODS Single-blind, randomized controlled trial. METHODS The study setting was the university hospital rehabilitation center. METHODS Fifty poststroke subacute hemiparetic patients with shoulder subluxation. METHODS Patients were randomly assigned into two groups. The position-triggered ES group received 30-minute ES sessions, 5 days per week for 3 weeks with specially modified Novastim® CU-FS1 (CU Medical Systems, Inc., Gangwon-do, South Korea) for motion triggering. The passive ES group received the same protocol without motion triggering. The vertical distance (VD) and the joint distance (JD), relative VD and JD (rVD, rJD), upper extremity component of Fugl-Meyer Motor Assessment (FMA<inf>upper</inf>), Motricity Index (MI), Manual Function Test (MFT), and peak torque of affected shoulder abductor (PT) were assessed at baseline (T0), end of electrical stimulation session (T1), and 3 weeks (T2) after treatment. RESULTS Repeated-measures analysis of variance revealed significant interaction between time and intervention on JD and rJD, indicating that shoulder subluxation was significantly more reduced in position-triggered ES than in passive ES (P<0.05). However, FMA<inf>upper</inf>, MI, MFT, and PT did not show this significance. The change of (∆)JD, ∆rVD, and ∆rJD in the motion-triggered ES group improved significantly more at T1 than in the passive ES group (P<0.05). This significant improvement was not seen at T2. CONCLUSIONS Position-triggered ES may be more effective than passive ES in improving poststroke shoulder subluxation; however, this effect was not maintained after the withdrawal of stimulation. CONCLUSIONS Position-triggered ES may be useful to reducing poststroke shoulder subluxation.