2012-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/701229摘要：糖尿病是一個具有高盛行率的慢性疾病。如果未及時予與適當的處置或治療，將導致相當嚴重的併發症與後果。因此，如何尋得良好的方法以促進糖尿病患者的代謝控制便成為全世界健康政策制定者的重要課題。持續照顧在健康照護體系之中具有重要的意義，尤其是針對糖尿病患者的照顧。在國外已有許多的文獻指出持續照顧對醫療照護結果具有舉足輕重的地位。但在台灣相關的研究卻相當稀少。此外，持續照顧對醫療照護結果的影響機制為何，則仍未明。本研究計畫之目的擬針對台灣新發病之糖尿病患者，探討持續照顧與藥物遵從性的關係，並進一步探討持續照顧是否藉由藥物遵從性達到影響醫療照護結果的作用。研究將利用國民健康保險資料庫之資料，採7 年追蹤之研究設計分析，將新診斷的糖尿病患者依據其持續照顧之分布區分為三組。每位患者將計算其藥物獲取率(medication possession rate)，並將藥物獲取率超過80%定義為具有良好的藥物遵從性。主要的醫療照護結果指標將包括每年的住院率及急診到訪率。因考慮有些住院或急診到訪與糖尿病患者之持續照顧並無直接關係，吾人將排除因意外、中毒或化學治療等因素所致之住院或急診到訪。主要的干擾因素控制將包括年齡、性別、健康狀態、藥物使用及其頻率、參與”按成效給付”(pay-for-performance)方案、時間效應及隨機受試者效應(random subject effect)等。本研究將可釐清台灣之糖尿病患者在持續照顧、藥物遵從性與醫療照護結果之間的關係。其結果將可提供健康主管機關在制定相關醫療政策的有力證據。<br> Abstract: Diabetes mellitus is a prevalent chronic disease with significant complicationsand serious consequences. Finding ways to promote better control of diabetes is animportant issue for health policy makers worldwide. Continuity of care (COC) is asignificant hallmark of a healthcare delivery system, which is particularly importantfor patients with diabetes. The effects of COC on health outcomes are well establishedin some previous publications from other countries. However, such a relationship israrely studied in Taiwan. Furthermore, the mechanism of the association is not fullyunderstood.The proposal of this study aims to examine the relationship between COC andmedication adherence and to investigate the mediating effect of medication adherenceon the association between COC and health outcomes for patients with newlydiagnosed diabetes in Taiwan.This study will employ a longitudinal design with a 7-year follow-up. Studyparticipants will be enrolled in a compulsory universal health insurance program inTaiwan, namely the National Health Insurance. Patients with newly diagnoseddiabetes will be categorized into three groups by the distribution of COC scores.We will use a propensity score-matching approach to place patients into threeCOC groups to increase the comparability among groups. A medication possessionrate (MPR) equal to or higher than 80% will be defined as good adherence. The mainhealthcare outcomes measures will be whether the subject had a hospital admissionfor any condition, or had an emergency department (ED) visit in each year during thestudy period. In addition, we will consider the possibility that some hospitaladmissions or ED visits have nothing to do with one’s level of continuity of outpatientcare; therefore we will exclude hospital admissions or ED visits with diagnoses of aninjury, poisoning (International Classification of Disease (ICD-9-CM) codes:850~995), or supplementary classifications (V-codes), such as chemotherapy.Generalized estimating equations will be used to control for age, sex, health status,intensity of the drug regimen, enrollment into a pay-for-performance program, timeeffects, and random subject effects.A clarification of the relationship among COC, medication adherence and healthoutcomes in the diabetes patients will provide important evidence for improving theCOC for diabetes patients, which may result in higher medication adherence andbetter health outcomes. The results will be helpful to indicate a direction that healthauthorities can follow.