口服降血糖藥失效之第2型糖尿病病人的疾病感受、胰島素治療信念、與胰島素治療意向

Abstract

口服降血糖藥失效的第2型糖尿病病人延遲胰島素注射，除了會造成長期血糖控制不良，增加罹患糖尿病合併症的風險，也影響生活品質，病人延遲胰島素治療常與病人的治療決策有關，本研究以常識模式為架構，探討糖尿病疾病感受與胰島素治療信念對胰島素治療意向的預測效果。 本研究為橫斷式比較性研究，樣本採立意取樣，以某醫學中心糖尿病門診共101位口服降血糖藥失效的第2型糖尿病病人為研究對象，回應率為88.59%，疾病感受與胰島素治療信念分別以自填式的糖尿病疾病感受量表與胰島素治療評價量表測量，胰島素治療意向則是請病人指出，在醫師持續建議胰島素治療的情況下，何時願意開始治療為測量方式，全部資料收集流程在實際接受胰島素治療之前完成，胰島素治療意向採兩種分類方式，第一種分類方式是選擇在不同時間開始胰島素治療者定義為意向組(44人)，無法決定何時開始胰島素治療者定義為缺乏意向組(57人)，第二種分類方式為選擇現在開始胰島素治療的病人獨立為一組(35人)，另一組仍然是缺乏意向組(57人)，疾病感受變項、胰島素治療信念與胰島素治療意向之間的關係以二元邏輯斯迴歸分析，以操作特性曲線(ROC)及曲線下面積來評值迴歸模式的預測效果。 研究對象平均年齡55.86歲(±11.47)，糖尿病罹病期平均為9.67(±6.62)年，糖化血色素平均為9.15%(±2.01%)，胰島素延遲期平均值7.05個月(±10.94月)，大部份(79.3%)的個案至少出現一種的糖尿病合併症。 依據二元邏輯斯迴歸結果，以胰島素治療意向第一種分組方式時，在控制糖化血色素值與糖尿病合併症後，疾病後果感受(OR=4.37, 95%CI=1.35-14.14, p=0.01)與胰島素正向治療信念(OR=5.58, 95%CI=1.54-22.27, p=0.00)是胰島素治療意向的促進因子，可以預測病人有意向開始胰島素治療，口服降血糖藥治療控制信念(OR=0.32, 95%CI=0.11-0.91, p=0.03)與胰島素負向治療信念(OR=0.18, 95%CI=0.04-0.73, p=0.01)是胰島素治療意向的阻礙因子，可以預測病人缺乏胰島素治療意向，整體模式的ROC曲線下面積為0.90。胰島素治療意向第二種分組方式，在控制糖化血色素值與年齡後，疾病後果感受(OR=4.50, 95%CI=1.04-19.41, p=0.04)與胰島素正向治療信念(OR=8.68, 95%CI=1.56-48.38, p=0.01)是胰島素治療意向的促進因子，可以預測病人有意向開始胰島素治療，口服降血糖藥治療控制信念(OR=0.19, 95%CI=0.05-0.74, p=0.01)與胰島素負向治療信念(OR=0.09, 95%CI=0.01-0.53, p=0.00)是胰島素治療意向的阻礙因子，可以預測病人缺乏胰島素治療意向，整體模式的ROC曲線下面積為0.93。 本研究發現顯示口服降血糖藥失效的病人對於胰島素治療意向的決策會依循自我調節的過程，病人透過疾病感受的回饋來衡量是否有必要繼續口服降血糖藥治療，當研究對象經歷到較多的自覺症狀，會激發疾病後果嚴重性的感受，病人意識到較嚴重的疾病後果並經驗到較強的情緒表徵，又覺得現有的口服降血糖藥控制效果較低時，就會對現有治療產生改變的想法，應用胰島素治療信念，當病人感受到胰島素治療符合疾病控制的需要，有較強的胰島素正向治療信念時，接受新治療的意願就會增加，即使對執行胰島素治療仍然感到有些困難或障礙。 本研究發現結合疾病感受與治療信念可以解釋第2型糖尿病病人在口服降血糖藥失效時，對於胰島素治療意向的決策過程，也提供實證依據以發展並設計協助病人接受胰島素治療的具體策略。

Delays in initiation of insulin therapy have been associated with poorer outcome in patients with failure to oral medicines. This study explores illness perceptions and treatment belief as predictors of intentions to start insulin therapy in a sample of patient with failure to oral drug. The utility of the self-regulation model of common sense of illness and treatment beliefs were examined in predicting the intentions to start insulin therapy. This study was cross-sectional comparative research with purposive sampling. A 88.59% response rate was obtained from a sample of 101 patients with failure to oral drugs in a medical center. Illness perceptions and beliefs towards insulin therapy were measured using the llness perception questionnaire-revised (IPQ-R-Diabetes) and the Insulin Treatment Appraisal Scale (ITAS). Intention to insulin was determined by asking participants to indicate when to start insulin therapy for the options of immediately, 3-months later, 6-months later, 9-months later, or cannot make decision if their physician continually recommends. The data collection was finished before the insulin therapy was acturally started. Two classifications for insulin intention were adopted. The first one was that those who cannot make decision about when to start insulin were defined as the subjects without intention (n=57). All others with other catergories of options were defined as subjects with intention (n=44). The second one was that those who choose to start insulin immediately (n=35) was defined as one group and the other group was also those who cannot make decision about when to start insulin (n=57). The relation between illness perceptions and beliefs towards insulin and the intention to insulin therapy was studied using the logistic regression. Then receiver operating characteristic (ROC) curves were used to evaluate the predicting capacity of regression model. Subjects’ mean age was 55.86 (±11.47) years. Average duration of known diabetes was 9.67 (±6.62) years with a mean HbA1c of 9.15 (±2.01) %. The number of months subjects delayed to insulin was 9.67(±10.94) months. The majority had diabetes-related complications (79.3%). After adjusted the HbA1c and the diabetic complication, binary logistic regression analysis for the first classification of insulin intention showed that consequences perceptions (OR=4.37, 95%CI=1.35-14.14, p=0.01) and the positive appraisal of insulin (OR=5.58, 95%CI=1.54-22.27, p=0.00) were promotive factors that could predict intention to insulin therapy after controlling the HbA1c and the diabetic complication. Control belief for oral medicine (OR=0.32, 95%CI=0.11-0.91, p=0.03) and the negative appraisal of insulin therapy (OR=0.18, 95%CI=0.04-0.73, p=0.01) were barriers that could predict lack of intention to insulin therapy. The area under the ROC curves of the logistic regression model was 0.90. After adjusted the HbA1c and age, binary logistic regression analysis for the second classification of insulin intention showed that consequences perceptions (OR=4.50, 95%CI=1.04-19.41, p=0.04) and the positive appraisal of insulin (OR=8.68, 95%CI==1.56-48.38, p=0.01) were promotive factors that could predict intention to insulin therapy. Control belief for oral medicine (OR=0.19, 95%CI=0.05-0.74, p=0.01) and the negative appraisal of insulin therapy (OR=0.09, 95%CI=0.01-0.53, p=0.00) were barriers that could predict lack of intention to insulin therapy. The area under the ROC curves of the logistic regression model was 0.93. These findings showed that the treatment decision making for subjects with failure to oral medicine would follow the self-regulation process. Subjects could evaluate the necessity to continue the oral medicine by the feedback form the illness perceptions. If subject experienced higher consequences perceptions which might by induced by symptoms related to diabetes and more intense emotion representation, and feel disappointed about the effectiveness of oral medicine, then he/she might have intention to change therapy. With the beliefs toward insulin, if subject perceived that the insulin will meet the disease needs, thean positive appraisal of insulin would dominate and he/she will have intention to start insulin even though some barriers to insulin injection remain. The results provides valuable insight into reasons why subjects may or may not start insulin therapy and have important implications for developing an intervention to promote the transition process from oral medicine to insulin injection.