[ 摘要 ]
背景：糖尿病共同照護網在台灣已推行多年，近期研究顯示，照護網介入在短期改善糖化血色素值及提高相關檢查指標執行率有良好成效，但在長期成效及影響因素的探討卻相對缺乏。台東地區人口組成多元，老年人口與原住民比率較高，糖尿病盛行率與十萬人口死亡率，均高於台灣整體，而糖尿病照護成效的相關探討仍欠缺實證的研究。
目的：了解台東某區域醫院糖尿病照護成效的分佈情形，與探討影響糖尿病照護成效的因素。方法：以台東某區域醫院糖尿病患為對象，在2003年1月至2004年12月間加入糖尿病共同照護網，並完成四次階段性追蹤的242位第2型糖尿病患進行分析，就其三年過程面檢查指標的執行與結果面糖化血色素的控制情形，作為糖尿病照護成效的評量與相關影響因素的探討。
結果：(一)糖尿病照護成效：過程面五項檢查指標(糖化血色素、血脂肪、血清肌酐酸、尿液檢查及眼底檢查)符合ADA照護指引建議，相對於國內其他研究有較高的達成比率。在結果面指標上，照護網介入後第一年的糖化血色素值，顯著低於首次登錄的基準值(7.93%vs.9.12 %)，後續第二、三年的糖化血色素值有逐漸上升的情形(8.20 % vs.8.27 %)，但相較於基準點仍具顯著的改善成效，其中首次糖化血色素偏高者改善成效相對較好；另外，糖化血色素控制不佳的比率(HbA1c>9%)在基準點為45.5%，第一年則顯著降低至23.1%，第二、三年控制不佳的比率則無顯著上升的趨勢 (28.5% vs.26.1%)。(二)影響糖尿病照護成效的因素：照護過程的影響因素集中在個案罹病狀況上，罹病期較長、有糖尿病合併症、相關藥物使用者，有較高的檢查次數；而年齡小於65歲、原住民、首次糖化血色素較高、糖尿病罹病期較長者，相較於非該特性者，其三年度糖化血色素值較高。
結論：本研究提供較長期的照護成效及其影響因素之探討，也證實糖尿病共同照護網的介入對於個案糖化血色素的控制有正面的成效；同時，在收案時相關因素包含：年齡、種族、首次糖化血色素值及糖尿病罹病期，都會是影響糖尿病照護成效的因素，可提供台東地區未來擬定糖尿病照護品質改善的參考依據。

[ 英文摘要 ]
Background: In Taiwan, the diabetes shared-care program has developed for many years. Recent researches showed the intervention can be effective way of improving glycemic control and increasing frequency of performance of relevant laboratory tests. But data on long-term quality of diabetes care are lacking. Among Taitung population, there are both the greatest proportion of aborigines and aged people and higher prevalence and mortality comparing to the rest parts of Taiwan. However there is few researches addressing on efficacy and associated factors of diabetes care among Taitung population.
Objective: To assess the efficacy and associated factors of diabetes shared-care program for type 2 diabetic patients among Taitung population.
Methods: In the retrospective study we enrolled 242 type 2 diabetic patients of a Taitung regional hospital, who participated in diabetes shared-care program and completed 4 follow-up times from January 2003 to December 2004. Data of demographics, outcome indicator: mean HbA1c value , and performance indicators : frequency of performance in HbA1c, lipid profile, serum creatinine, urine analysis and fundus exam were collected within the 3-year period.
Results:(1) There is higher frequency of five performance indicators than that of previous researches in Taiwan. The greatest reduction in mean HbA1c value was detected during the first follow-up year, then it is progressively increasing by time (baseline,1st,2nd ,3rd year is 9.12%,7.93%,8.20% ,8.27%). The outcome indicator is still significantly improved at the end of a 3-year period. (2) The longer duration, comorbidity and relative medication used had higher frequency of performance indicators. Also aged < 65 years old, aborigines, higher initial mean HbA1c value, longer duration might contributed to poor glycemic control.
Conclusions: Our study demonstrated the long-term efficacy of diabetes shared- care program and baseline characteristic (age , ethnic , initial HbA1c value and duration ) were associated factors of diabetes care. These findings can help evaluate the quality of diabetes-care among Taitung population.