病人積極度測量之發展，及其與病人行為間關係之研究-以糖尿病病人為例

Abstract

背景與動機：「病人」為病人安全之核心角色，欲避免醫療疏失的最佳方法就是病人在醫療團隊裡擔任一位積極的成員，許多研究指出病人應該積極參與醫療決策與自我照顧，故了解「病人積極度」對於當前病人安全議題十分重要。 對於病人積極度的討論，現存的理論與方法中已就不同層面的積極度進行測量，例如健康人格控制、自我管理行為、自我效能、健康相關行為等，但是這些測量都傾向著重於單一行為的預測，並未包括慢性病人所需的知識、技能、信念與行為等較廣泛因素的積極度測量。 由於國內尚無相關衡量「病人積極度」的測量工具與應用之相關研究，故本研究引用由美國奧勒岡大學Professor Hibbard等學者發展並已加以應用的病人積極度(patient activation measure)測量。研究目的為發展中文化的病人積極度測量工具，以及探討病人積極度與糖尿病病人行為間之關係。 研究方法：本研究獲得原作者Professor Hibbard的授權，中文化過程包括英翻中、中翻英之回譯，並邀請Professor Hibbard作為回譯版本翻譯品質標準，依其給分進行修正，若該題項較低分則重新修正中文內容，最後依前測結果修改後完成定稿。 本研究採橫斷性研究，在徵求台北市有參與全民健康保險糖尿病醫療服務改善方案試辦計劃醫院合作後，共計有4家醫院同意參與本研究，包括1家醫學中心、3家區域教學醫院，以其中照護之糖尿病病人作為分析單位。本研究採立意抽樣，於2005/5/16至2005/6/15收案期間共計收集277位有效樣本。 結果：本研究中病人積極度工具共有22題具有高度內部一致性(Cronbach α係數為0.882)，病人行為題目計有13題內部一致性尚可(Cronbach α係數值為0.696)；使用探索性因素分析結果發現病人行為量表萃取出四個因素，包括遵從性行為、控制行為、自我照護行為與飲食行為等，經複迴歸分析，發現知識總分、醫護人員支持積極程度與婚姻狀況與病人積極度有顯著相關。 有許多控制變項與病人行為有顯著相關，包括性別、教育程度、居住現況、年齡，糖尿病病史、其他慢性病症、糖尿病類型、治療方法、是否納入健保局糖尿病試辦計畫、個案醫院醫院、醫護人員支持病人積極程度與糖尿病知識得分。 在控制其他變項下，本研究發現「病人積極度」與「病人行為」有正向關係。

Objectives: Patient plays a key role in patient safety. The best way to prevent medical error is that patient becomes an active member in the practice team. Several studies mentioned that patients should be active to participate in medical decision and self care. It is very important to understand “Patient Activation Measurement” regarding the topic of patient safety.

In existing theory and method, some instruments were applied to measures health locus control, self-management behavior, self-efficiency, health-related behavior etc, but these measures tend to focus on the prediction of a single behavior, not including the necessary knowledge, skill, confidence, and behavior of chronic patients.

Owing to lack of available instrument in Taiwan, this study adopted Patient Activation Measurement (PAM), developed by professor Hibbard etc, in Oregon University. The first purpose of this study is to develop Chinese version instrument of patient activation measurement. The second purpose is to explore the relationship between PAM and behavior for diabetes patients. Method: This study obtained the authorization directly from professor Hibbard. The modification procedures included translation from English to Chinese, back- translation from Chinese to English, rated of English by professor Hibbard as gold standard, and modify the Chinese if there is low rating in previous step. The final version was decided after pretest of the draft. This is a cross-sectional study. After searching for help from hospitals joined pay for performance program in the National Health Insurance, four hospitals agreed to cooperate with the study that included one medical center and three regional hospitals. Diabetes patients in sample hospitals were treated as unit of analysis. All samples came from convenient sampling. There were 277 valid samples with collection period from 2005/5/16 to 2005/6/15. Result: There are 22 questions in PAM instrument with high internal consistency (the coefficient of Cronbach α= 0.882). There are 13 question in patient’s behavior with moderate internal consistency (the coefficient of Cronbach α= 0.696). Exploratory factor analysis is carried out to abstract four constructs in Patient behavior including compliance behavior, control behavior, self-care behavior and diet behavior. From multiple regression analysis, the results found that diabetes knowledge, the degree of active support for patient from doctors and nurses and marital status had significant relationship with PAM. There are many factors with significant relationship with patient behavior that included gender, education level,inhabit situation age, history of diabetes, other chronic diseases, diabetes type, treatment plan, whether patients covered by the pay for performance program, hospital as a whole, the degree of active support for patient from doctors and nurses, an diabetes knowledge. After controlled for personal characteristics, the study found positive relationship between PAM and patient’s behavior.