目的：給藥時間錯誤為給藥錯誤的原因之一，但目前相關研究仍有限。本研究初探住院病人給藥時間錯誤的情形及常見的藥物種類，盼能作為後續改善給藥作業流程的政策參考。
方法：本研究利用某醫學中心住院條碼給藥系統進行回溯性分析，分別比較2015、2016 和2017 年4 月間給藥時間錯誤率及藥物種類。給藥時間錯誤定義為常規治療藥物給藥時間點前後超過60 分鐘，立即治療藥物給藥時間點則以醫師開方30 分鐘為限。研究利用anatomical therapeutic chemical (ATC) 碼統計給藥時間錯誤的藥物種類，並探討不同工作時段間給藥錯誤率。
結果： 研究共收錄2,605,276 筆住院病人給藥紀錄， 平均給藥時間錯誤率為10.9%， 但逐年下降（2015 年4 月：12.1%；2016 年4 月：10.4%；2017 年4 月：10.0%）。不同時段的給藥時間錯誤率則以大夜班時段最高 (0:00 AM ~ 8:00 AM; 22.3%)，其次為小夜班時段 (4:00 PM ~ 0:00 AM; 12.2%) 和白班時段 (8:00 AM ~ 4:00PM; 9.7%)。最常發生延遲或提早給藥的藥物種類為anilides 類解熱鎮痛藥（ATC 碼：NO2BE），占整體逾時藥物的5.7%，其次為促進腸胃蠕動劑（ATC 碼：A03FA；5.1%）和質子幫浦抑制劑（ATC 碼：A02BC；4.9%），這些藥物也均為各年給藥時間錯誤的藥物種類前三名。
結論：住院病人給藥時間錯誤率雖有逐漸下降的趨勢，但常發生給藥時間錯誤的藥物種類卻未有明顯變化，同時大夜班時段的給藥時間錯誤率偏高。後續研究仍須探討導致延遲或提早給藥的根本原因及病人後續影響。

Objective: Administration timing errors are common medication errors, but related studies are still lacking in Taiwan. This study focuses on the rates and drug classifications of administration timing errors in inpatients, and our findings could serve as hospital policy references regarding improvements in drug administration.
Methods: This retrospective study uses bar-code medication administration system (BCMA) records from April of 2015, 2016 and 2017 from the largest medical center in Taiwan. We define timing errors as the administration of drugs to patients earlier or later than 60 minutes for regular use cases, and later than 30 minutes for immediate use cases following doctors’ orders. The anatomical therapeutic chemical (ATC) codes is used to classify different agents given in wrong time. We also analyze the rates of timing errors at different times of the workshift.
Results: We analyze a total of 2,605,276 records from BCMA. The mean rate of administration timing errors is 10.9%, which decreases over time from 12.1% in April 2015, 10.4% in April 2016 to 10.0% in April 2017. We find the highest rate of timing errors occurring during the night shift (0:00 AM–8:00 AM; 22.3%), followed by the evening shift (4:00 PM–0:00 AM; 12.2%) and morning shift (8:00 AM–4:00 PM; 9.7%). The most common agents given at the wrong time include anilide analgesics (ATC: N02BE; 5.7%), propulsives (ATC: A03FA; 5.1%) and proton pump inhibitors (ATC: A02BC; 4.9%) which account for the top 3 drug classifications in each year.
Conclusions: Our study indicates that administration timing errors in inpatients are decreasing, but there are no changes in the most common drug classifications of such timing errors. Attention must be drawn to the higher rates of timing errors during the night shift. Future studies evaluating the
root causes of administration timing errors and their impact on patient care will be warranted.