以多層次分析探討個人和醫師層次特質對新診斷無併發症糖尿病病患發生相關併發症之影響

The influence of individual and physician level characteristics on the incidence of the diabetes relative complications of new diagnosed diabetic patients with no complications in multilevel analysis

劉佳鑫(Jia-Sin Liu)

指導教授 ： 李建宏

高雄醫學大學/健康科學院/公共衛生學研究所/碩士(2008年)

https://doi.org/10.6832/KMU.2008.00080

摘要

本研究探討影響糖尿病病患發生糖尿病相關併發症的個人層次與醫師層次因素。研究收集具有代表性的全民健保資料，在2000年選取15,954位新診斷無併發症糖尿病病患，以及利用年齡性別1:1頻率匹配的選取人數相同的對照。研究追蹤到糖尿病病患與對照2000年到2004年慢性腎臟疾病、心臟血管疾病、腦中風、周邊動脈血管疾病、截肢與視網膜病變等糖尿病相關併發症發生情形。
在平均4.5年的追蹤時間下，新診斷無併發症糖尿病病患發生慢性腎臟疾病、心臟血管疾病、腦中風與截肢的風險顯著高於對照世代，發生率比(relative risk, RR)分別為2.3、1.2、1.2與2.0，95％信賴區間為2.2-2.5、1.1-1.3、1.1-1.3、1.8-2.1。已有高血壓或血脂異常的新診斷無併發症糖尿病病患在心臟血管疾病、腦中風、周邊動脈血管疾病與視網膜病變皆顯著低於的同樣狀態的對照世代，RR分別為0.5、0.6、0.5與0.5，95％信賴區間為0.4-0.5、0.5-0.7、0.4-0.5與0.5-0.4。高血壓與血脂異常會修飾糖尿病與相關併發症的關係。在multilevel logistic regression model中控制糖尿病患者的個人層次與醫師層次因素後，發現糖尿病患者的個人層次因素，年齡較高、有高血壓、換醫師次數較多皆會有顯著高的慢性腎臟疾病、心臟血管疾病、腦中風與截肢的發生比例；醫師層次方面，醫師所在醫師機構的層級較低（如基層診所）與醫師所在地區的人口密度較低的情況皆會有顯著高糖尿病病患慢性腎臟疾病、心臟血管疾病、腦中風與截肢發生比例。
糖尿病與相關併發症的關係會受到高血壓與血脂異常的影響。個人年齡、高血壓、血脂異常與換醫師行為、病患醫師所在視機構層級與所在地區人口密度會影響糖尿病患者的慢性腎臟疾病、心臟血管疾病、腦中風與截肢發生情形。

關鍵字

多層次分析 ； 糖尿病 ； 併發症 ； 健保資料庫

並列摘要

This study investigated the influence of individual and physician level characteristics on the incidence of the diabetes relative complications on new diagnosed diabetic patients with no complications. A cohort consisting of 15,954 new diagnosed diabetic patients with no complications and their age and gender matched non-diabetes controls were identified from the database of Taiwan National Health Insurance in 2000. Study participants were followed up annually until 2004.
On an average of 4.5-years of follow-up, diabetic patients were observed to have 2.4, 1.2, 1.2 and 2.0-fold higher risk of contracting chronic kidney disease, coronary artery disease, stroke and amputation than the controls. The 95% confidence intervals were 2.2-2.5, 1.1-1.3, 1.1-1.3 and 1.8-2.1, respectively. On the other hand, diabetes patients with hypertensive or dyslipidemia were observed to have 0.5, 0.6, 0.5 and 0.5 -fold higher risk of contracting chronic kidney disease, coronary artery disease, stroke and amputation than the controls. The 95% confidence intervals were 0.4-0.5, 0.5-0.7, 0.4-0.5 and 0.5-0.4, respectively. The multilevel logistic regression model was used to evaluate the effect of individual and physician characteristics associated with the incidence of diabetes relative complications on diabetic patients. In the individual level characteristics of diabetic patients, age, hypertension, and visit physician change were shown to be significant risk factors for chronic kidney disease, coronary artery disease, stroke and amputation on diabetes patients. In the physician level characteristics of diabetic patients, hospital type in clinic and low density of population were shown to be significant risk factors for chronic kidney disease, coronary artery disease, stroke and amputation on diabetes patients.
The associate or diabetes and diabetes relative complications were modified by patients with or without hypertensive and dyslipidemia. The incidence of chronic kidney disease, coronary artery disease, stroke and amputation on diabetic patients were influenced by age increasing, hypertension, visit physician change more, hospital type in clinic and low density of population.

並列關鍵字

multilevel analysis ； multilevel logistic ； diabete ； complications ； National Health Insurance database

參考文獻

1. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes care. 2004;27(5):1047-1053.

2. 魏榮男, 莊立民, 林瑞雄, 趙嘉玲, 宋鴻樟. 1996~2000年台灣地區糖尿病盛行率與住院率. 臺灣公共衛生雜誌. 2002;21(3):173-180.

3. Roglic G, Unwin N, Bennett PH, Mathers C, Tuomilehto J, Nag S, Connolly V, King H. The burden of mortality attributable to diabetes: realistic estimates for the year 2000. Diabetes care. 2005;28(9):2130-2135.

4. 民國95年衛生統計系列(一)死因統計. 台北: 行政院衛生署; 2008.

5. Stamler J, Daviglus ML, Garside DB, Dyer AR, Greenland P, Neaton JD. Relationship of baseline serum cholesterol levels in 3 large cohorts of younger men to long-term coronary, cardiovascular, and all-cause mortality and to longevity. JAMA. 2000;284(3):311-318.