醫師使用電子病歷交換的影響因素及效益評估─以健保雲端藥歷系統為例

Influential factors and benefits of physicians using health information exchange – using National Health Insurance PharmaCloud System as an example

莊秋華(Chiou-Hwa Chuang)

指導教授 ： 楊銘欽

國立臺灣大學/公共衛生學院/健康政策與管理研究所/博士(2018年)

https://doi.org/10.6342/NTU201803078

摘要

目的：本研究欲分析影響醫師使用電子病歷交換的因素及效益評估，以健保雲端藥歷系統為例，採用科技持續理論 (Technology Continuance Theory, TCT)，瞭解醫師使用健保雲端藥歷系統的影響因素及其效益，即是否能降低病人的藥品項數、藥費及增加被發現重複用藥機率、用藥交互作用機率之情形，以作為健保署及醫療機構推動持續使用的參考。方法：依據研究目的，將研究分為兩部分，第一部分以科技持續理論為基礎，設計結構式問卷，調查台灣某醫療體系的醫師，該體系包括一家醫學中心、兩家區域醫院及三家地區醫院等，針對門診主治醫師在使用健保雲端藥歷系統，採用預載比對雲端藥歷和手動查詢雲端藥歷兩種方式，在「與期待相符」、「知覺有用性」、「知覺易用性」、「態度」、「滿意度」、「持續使用意向」等變項，使用結構方程模式(SEM)分析以驗證各變項間的因果關係。第二部分探討醫師使用健保雲端藥歷系統的效益評估，採用差異中之差異模型(DID)來分析，有簽署同意預載比對雲端藥歷病人是否能讓就診門診成年病人的每年每次藥品項數、每人每次的藥費有效降低及增加被發現重複用藥、用藥交互作用機率。結果：本研究共發出775份問卷，回收有效問卷528份(回收率68%)。根據醫師基本特質對變項的影響，本研究發現醫師基本特質對於預載比對雲端藥歷系統的看法，不同醫院層級填答者對預載比對雲端藥歷的「知覺易用性」的看法有顯著差異與不同性別填答者對預載比對雲端藥歷的「態度」的看法有顯著差異。另，醫師年齡越低、服務年資越短，其對手動查詢雲端藥歷的「與期望相符」、「知覺有用性」、「知覺易用性」、「態度」、「滿意度」、「持續使用意向」的得分越高，且不同醫院層級對手動查詢雲端藥歷的「與期望相符」、「知覺有用性」「知覺易用性」、「態度」、「滿意度」、「持續使用意向」的看法會有顯著差異；而每次平均門診人數較少的醫師，其對手動查詢雲端藥歷的「知覺有用性」、「知覺易用性」、「態度」的看法和平均門診人數較多的醫師不同，且「滿意度」和「持續使用意向」得分越高。路徑分析之結果顯示，會影響醫師對預載比對雲端藥歷的「持續使用意向」的因素包括「認知有用性」(p<0.0001)、「滿意度」(p<0.001)、「態度」(p<0.001)；但影響醫師手動查詢雲端藥歷「持續使用意向」的因素只有「滿意度」(p<0.001)和「態度」(p<0.001)。「與期望相符」與「知覺易用性」在預載比對雲端藥歷和手動查詢雲端藥歷上，都會正向影響「知覺有用性」(p<0.001, p<0.001)；「知覺有用性」和「知覺易用性」兩變項在預載比對和手動查詢雲端藥歷上，也都會正向影響「態度」(p<0.001, p<0.001)；但「知覺有用性」對「滿意度」的影響只有在預載比對雲端藥歷上顯著。健保雲端藥歷查詢效益評估分析上，在控制性別、年齡、三高及思覺失調等疾病病人特質後，病人有簽署同意預載比對雲端藥歷同意書之後測發現平均每人每次藥品項次減少7.0%（p<0.001）、平均每人藥費則未顯著增加。另外，被發現有重複用藥的機率顯著較高，是未簽署者的1.83倍（p<0.0001）；被發現有用藥交互作用的機率和未簽署者之差異則未達統計顯著。結論：當醫師對健保雲端藥歷系統的「期望」與實際使用的相符性越接近，則其對健保雲端藥歷系統的「知覺有用性」越高，自然而然地對該系統的「滿意度」和「態度」就越良好，且對「持續使用」健保雲端藥歷系統所造成的正向影響則會越強，持續使用健保雲端藥歷系統的時間則會越久。病人有簽署同意預載比對雲端藥歷之後，顯著的影響病人門診的藥品項數降低，但門診藥費卻沒有減少。另外，被發現有重複用藥的機率顯著較高，但被發現有用藥交互作用之機率則没有差異。

關鍵字

雲端藥歷 ； 電子病歷交換 ； 電子病歷 ； 科技持續理論 ； 科技接受模式 ； 雲端資訊系統 ； 醫療資訊系統

並列摘要

Objectives: This research aims to use Technology Continuance Theory (TCT) to investigate the factors influence physicians using pharmaCloud system and to evaluate its benefits, which includes whether the number and cost of prescribed medications are redueced and the probability of discovering repetitive medication and medication interaction are increased. Method: According to the research objective, the research consists of two parts: the first part is to construct a structured questionnaire that was based on the TCT, which was used in order to investigate physicians in a medical system, which consists of a medical center, two regional hospitals, and three district hospitals in Taiwan. The research focuses on investigating the “confirmation”, “perceived usefulness“, “perceived ease of use“, “attitude“, “satisfaction“ and “continuance usage intention“ towards preload-based comparison and manual search in pharmaCloud for attending physicians during their outpatient clinics. The research uses path analysis to analyze the cause and effect relationship between variables. The second part of the research evaluated the benefits of physicians using the Healthcare pharmaCloud system. The research uses Difference in Differences (DID) to analyze whether patients who signed the consent form to use preload-based comparison pharmaCloud experienced reduced numbers and cost of prescribed medication and an increase of the probability of discovering repetitive medication and medication interaction than those who did not sign the consent form. Results: The research sent out 775 copies of questionnaires, and collected 528 copies, with a response rate of 68%. In terms of the effects of physicians’ characteristic on using preload-based pharmaCloud, the research found that physicians from different levels of hospitals have significantly different opinions toward the “perceived ease of use” in preload-based comparison in pharmaCloud. Gender of the physicians also significantly affected their attitude towards preload-based comparison in pharmaCloud. In terms of manual search in pharmaCloud, the research found that in regards to “perceived ease of use”, “attitude”, “satisfaction” and “continuance usage intention”, physicians who are younger and have fewer years of service, scored higher in “confirmation”, “perceived usefulness”, “perceived ease of use”, “attitude”, “satisfaction” and “continuance usage intention” towards manual search in pharmaCloud. In addition, levels of hospitals also significantly affect physicians’ opinions toward “confirmation”, “perceived usefulness”, “perceived ease of use”, “attitude”, “satisfaction” and “continuance usage intention” towards manual search in pharmaCloud. Physicians who had less average number of outpatients seen per clinic session scored higher in “satisfaction” and “continuance usage intention” and their opinions in “perceived usefulness”, “perceived ease of use” and “attitude” towards manual search in pharmaCloud is significantly different from that of physicians who had more average number of outpatients seen per clinic session. The results of path analysis show that factors affect physicians continue to use preload-based comparison in pharmaCloud included “perceived usefulness” (p<0.001), “satisfaction” (p<0.001), and “attitude” (p<0.001). However, factors that influence physicians’ continue to use manual search in pharmaCloud are only “satisfaction” (p<0.001) and “attitude” (p<0.001). “Confirmation“ and “perceived ease of use“ both affect physicians’ “perceived usefulness“ towards preload-based comparison in pharmaCloud (p<0.001, p<0.001, respectively), and both “perceived usefulness” and “perceived ease of use” positively affect “attitude” (p<0.001, p<0.001, respectively) on both preload-based comparison and manual search in pharmaCloud. However, the effects of “perceived usefulness” and “perceived ease of use” on “satisfaction” can only be seen in preload-based comparison in pharmaCloud. In terms of the benefits of the healthcare pharmaCloud system, the research has controlled patients’ gender, age, triple H (hypertension, hyperglycemia, and hyperlipidemia) and Schizophrenia and found that by signing consent for preload-based comparison in pharmaCloud, the number of prescribed medications has decreased by 7% (p<0.001) per visit per person, but the costs of prescribed medication per visit per person did not increase significantly. The probability of discovering repetitive medication was 1.83 times higher than that of outpatients who did not sign the consent form while the probability of discovering medication interaction was not significantly different.
Conclusion: When physicians’ actual use of pharmaCloud has met their expectation, they will have higher levels of confirmation, better perceived usefulness, and it will naturally increase their satisfaction and attitude towards pharmaCloud and positively affect them to continue to use pharmaCloud. After outpatients have signed the consent form for preload-based comparison on pharmaCloud, the number of prescribed medications decreased significantly while the change of cost of prescribed medication was not significant. In addition, the probability of discovering repetitive medication increased significantly while the probability of discovering medication interaction did not change significantly.

並列關鍵字

PharmaCloud ； Health Information Exchange (HIE) ； Electronic Medical Records ； Technology Continuance Theory (TCT) ； Technology Acceptance Model (TAM) ； MediCloud System, Health Information Systems (HIS)

參考文獻

陳福基, 蕭世榮, 陳啟元, & 杜素珍. (2005). 影響醫院接受行動護理站因素之研究-以南部某區域教學醫院為例. 資訊管理學報, 12(S), 67-89.

Davis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Quarterly, 13(3), 319-340. doi:10.2307/249008

Abdekhoda, M., Ahmadi, M., Gohari, M., & Noruzi, A. (2015). The effects of organizational contextual factors on physicians' attitude toward adoption of Electronic Medical Records. J Biomed Inform, 53, 174-179. doi:10.1016/j.jbi.2014.10.008

Abramson, E. L., McGinnis, S., Edwards, A., Maniccia, D. M., Moore, J., Kaushal, R., & Investigators, H. (2012). Electronic health record adoption and health information exchange among hospitals in New York State. J Eval Clin Pract, 18(6), 1156-1162. doi:10.1111/j.1365-2753.2011.01755.x

Adler-Milstein, J., Bates, D. W., & Jha, A. K. (2011). A survey of health information exchange organizations in the United States: implications for meaningful use. Ann Intern Med, 154(10), 666-671. doi:10.7326/0003-4819-154-10-201105170-00006