商業智慧工具Power Query應用於醫院指標收集分析之成效評估

Performance Evaluation of Using Power Query in Hospital Indicator Collection and Analysis

曾祐琳(Yu-Lin Tseng)

《醫務管理期刊》 22卷4期 (2021/12) Pp. 283-297

https://doi.org/10.6174/JHM.202112_22(4).283

摘要

目的：為保證臨床服務品質與持續改善內部流程，指標監測是當今醫院品管工具中不可或缺的一環，但非資訊專業人員面對雜亂的巨量資料時，作業往往耗時費力，故嘗試導入商業智慧（Business Intelligence, BI）大數據處理工具，盼提升醫院行政效率。方法：本研究運用Excel之Power Query增益集，以門診十項重要檢驗（查）相同病人28日內再執行率之指標收集分析流程為研究對象，建置資料處理模型，設計自動化流程取代人工作業。結果：該指標處理流程自959分鐘降至18分鐘，節省約98%作業時間。推估50至200指標項目階段性導入Power Query後，人事成本一年約節省25萬至150萬元。結論：本研究成果除提升作業效率，兼具減少隨機錯誤、統計條件調整便利、延長電腦使用效能與具備橫向運用之附加效益。

關鍵字

商業智慧 ； 醫院指標 ； 大數據資料處理工具 ； Power Query ； 自動化流程

並列摘要

Objectives: Indicator surveillance is critical in hospital quality management to uphold clinic service quality and continue improving internal processes. However, substantial time is required for non-information technology personnel to collect and analyze indicators because of dirty data and various data formats. This study aims to evaluate the effectiveness of introducing a business intelligence big data analysis tool to improve administrative efficiency. Methods: This study uses a user-friendly and accessible Excel Power Query addin to create a data processing model to import, normalize, and integrate data from various sources. The indicators include 10 critical examinations for the same outpatient reevaluated for the same examination within 28 days. This model simplifies traditional tasks through process automation. Results: The data processing model performs excellently. It reduces work time by approximately 98%, from 959 minutes to 18 minutes. Additionally, according to an initial assessment conducted after having introduced this business intelligence method to 50-200 indicators surveillance, the hospital could reduce annual personnel costs by approximately NT$250,000 to NT$1,500,000. Conclusions: The use of a business intelligence analysis tool increases work efficiency. Additional benefits of automated processing include a decrease in random errors, quick adjustment of statistic conditions, longer-lasting computer performance, and reuse of models in similar data manipulation procedures.

並列關鍵字

business intelligence ； hospital indicators ； big data ； Power Query ； process automation

參考文獻

Abràmoff, M. D., Lavin, P. T., Birch, M., Shah, N., & Folk, J. C. (2018). Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offi ces. npj Digital Medicine, 1(39), 1-8. https://doi.org/10.1038/s41746-018-0040-6

Ali-Ozkan, O., & Ouda, A. (2019). Key-based Reversible Data Masking for Business Intelligence Healthcare Analytics Platforms. ISNCC, 1-6. https://doi.org/10.1109/ISNCC.2019. 8909125.

Chen, X., Wang, L., Ding, J., & Thomas, N. (2016). Patient flow scheduling and capacity planning in a smart hospital environment. IEEE Access, 4, 135-148. https://doi.org/10.1109/ACCESS.2015.2509013

Jamil, F., Hang, L., Kim, K., & Kim, D. (2019). A Novel Medical Blockchain Model for Drug Supply Chain Integrity Management in a Smart Hospital. Electronics, 8, 505. https://doi.org/ 10.3390/electronics8050505

Khedr, A., Kholeif, S., & Saad, F. (2017). An Integrated Business Intelligence Framework for Healthcare Analytics. International Journal of Advanced Research in Computer Science and Software Engineering, 7, 263-270. https://doi.org/10.23956/ijarcsse/SV7I5/0163.