目的：院內中風因症狀不典型且流程冗長，預後常劣於院外個案。本專案透過PDCA循環建構「護理師引導制度」，強化早期偵測守門人角色，旨在優化救治時效與品質。方法：納入2019-2025年共479名個案（基線期51人，改善後428人）。P-D期建立N4護理師主導之雙軌支援，導入擬真演練與視覺評估看板；C-A期運用 Power BI監測斷點並滾動修正。以 t檢定、卡方檢定及標準差(SD)評估穩定性，並嚴格排除類中風個案。結果：發現至完成電腦斷層掃描時間由222分鐘縮短至85分鐘(p<.001)，2025年達66分鐘；標準差由72降至25，流程趨於穩定。再灌流治療率由2.4%升至11.4%(p<.05)；3個月良好功能恢復率(mRS 0–2)由22.2%提升至35.7% (p<.05)。統計證實兩組入院美國國家衛生研究院腦中風量表(NIHSS)、年齡及合併症均無顯著差異(p>.05)。結論：結合 PDCA之護理師引導制度能有效填補救治真空期。透過標準化訓練與即時監測，強化守門人價值，確保跨科別救治之穩定性與可複製性，具臨床推廣潛力。

Purpose:Patients with in-hospital stroke (IHS) often present with atypical symptoms, leading to delayed workfows and poorer outcomes compared with those experiencing strokes in community settings. This study established a nurse-led stroke assessment system integrated with the PDCA (Plan-Do-Check-Act) cycle to facilitated early-detection of IHS and improve treatment efciency and quality. Method:A total of 479 patients with IHS from 2019 to 2025 were included (baseline group: 51; post-intervention group: 428). During the Plan and Do phases, a dual-track support system led by level N4 nurses was established, incorporating high-fdelity simulations and visual assessment tools adapted from focused assessment with sonography for trauma (FAST). During the Check and Act phases, Microsoft Power BI was used to monitor process bottlenecks and support iterative refnement of the system. Statistical analyses included t tests and chi-square tests and standard deviation (SD) to assess process stability. Patients with stroke mimics were strictly excluded. Result: Following implementation, the mean onset to computed tomography (CT) time was reduced from 222 minutes to 85 minutes (p < .001), reaching 66 minutes in 2025. The SD decreased from 72 to 25, indicating enhanced process stability, and the immediate notifcation rate reached 100%. Reperfusion therapy rates increased from 2.4% to 11.4% (p < .05), and the 3-month favorable functional recovery rate (modifed Rankin scale score 0–2) increased from 22.2% to 35.7% (p < .05). Baseline characteristics, including admission National Institutes of Health Stroke Scale (NIHSS) score, age, and comorbidities, showed no signifcant diferences between groups (p > .05). Conclusion: The nurse-led stroke assessment system integrated with the PDCA cycle efectively addressed a critical gap in IHS care. Through standardized training and real time monitoring, this system enhanced nursing responsiveness and ensured the stability and replicability of stroke care across departments.