請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/42253
標題: | 社區雄性禿多階段進展和代謝症候群相關之研究 Community-based study on multi-step progression of androgenetic alopecia associated with metabolic syndrome |
作者: | Lin-Hui Su 蘇琳惠 |
指導教授: | 陳秀熙(Tony Hsiu-Hsi Chen) |
關鍵字: | 雄性禿,代謝症候群,社區,多階段,馬可夫鏈,貝式., androgenetic alopecia,metabolic syndrome,community,multi-state,Markov chain,Bayesian., |
出版年 : | 2011 |
學位: | 博士 |
摘要: | 背景:
代謝症候群與第二型糖尿病及心血管疾病有關,過去文獻中有一些針對雄性禿與代謝症候群有關因子的相關性研究,但是結果不盡相同。除此之外,雄性禿屬於一種漸進式進展之疾病,雄性禿進展速度以及其與代謝症候群之相關性也未曾被研究過;而且,對於這些議題之社區性研究,雄性禿的錯誤分組經常是一個需要關切的問題。因此,對於雄性禿進展速率以及其與代謝症候群之關係,尤其此關係是否為錯誤分組的影響,是非常值得研究的主題。 目的: (1) 控制其他干擾因子之後,釐清雄性禿與代謝症候群是否具有相關性。 (2) 在考慮或不考慮錯誤分組的情況下,評估雄性禿進展速率以及探討其與代謝症候群之相關性。 方法: 於台灣某社區進行族群為基礎之橫斷性調查,使用Norwood及Ludwig分類法針對男性型式及女性型式掉髮程度做評估,並收集與代謝症候群有關以及其他可能的危險因子之資料。(1) 於2005年四月至六月,共計740名40-91歲之男性參加研究,以評估雄性禿與代謝症候群之相關性,使用羅吉斯迴歸模式分析雄性禿與代謝症候群之相關性;(2) 於2005年共計4,633名女性及2,362名男性參與研究,於2010年共計25,118名女性及16,884名男性參與研究,其中899名女性及584名男性參與此兩次篩檢用於評估發生率,所有之資料用於評估雄性禿進展速率以及探討其與代謝症候群之相關性。使用多階段馬可夫模式分析,並應用貝式分析探討錯誤分組之影響。 結果: (1) 結果顯示當控制年齡、雄性禿之家族史、及抽菸後,雄性禿與代謝症候群具有統計上顯著之相關性(Odds ratio (OR) = 1.67, 95% CI: 1.01, 2.74),而且與代謝症候群五個項目之數目也有相關性存在(OR= 1.21, 95% CI: 1.03, 1.42)。在代謝症候群的項目中,高密度膽固醇與雄性禿最具相關性 (OR= 2.36, 95% CI: 1.41, 3.95, p= 0.001)。(2) 五年追蹤後89/745(12.0%)女性及58/369(15.7%)男性產生雄性禿,換算為2.4%及3.1%每人年之男、女性發生率。在雄性禿之進展速率上,當調整年齡、性別與雄性禿家族史之後,代謝症候群與第二階段進展有統計上顯著之相關性(Hazard ratio (HR) =1.16, 95%CI: 1.04, 1.30)、但與第一階段進展無關 (HR=1.03, 95%CI: 0.98, 1.08)。代謝症候群中某些個別因子與雄性禿進展具有相關性,包括高密度膽固醇與第一階段進展相關(HR=1.07, 95%CI: 1.01, 1.13)、血糖與糖尿病和前後兩階段進展皆有相關 (HR=1.06, 95%CI: 1.01, 1.11; HR=1.20, 95%CI: 1.08, 1.33)、高血壓和前後兩階段進展也有相關(HR=1.04, 95%CI: 1.01, 1.08; HR=1.16, 95%CI: 1.06, 1.26)。 結論: 雄性禿與代謝症候群存在具有統計上顯著之相關性,而且與代謝症候群條件符合之數目也有統計上顯著之相關性。關於雄性禿之進展速率上,代謝症候群與由輕度或中度進展至嚴重程度之雄性禿具有相關性,這個結果即使調整雄性禿錯誤分組之情況下也相同。這些結果顯示雄性禿與代謝症候群存在顯著之相關性,因此,針對中度或嚴重之雄性禿患者早期偵測代謝症候群,可以早期介入並可能減少日後心血管疾病與糖尿病之造成之風險與併發症,在我們的研究中,我們可以評估雄性禿錯誤分組之程度、並獲得調整後的雄性禿進展速率與代謝症候群之相關性,這些結果更可用於預測雄性禿進展之機率、並應用於未來風險預測及成本效益分析之研究。 Background: Several previous studies have investigated the association between androgenetic alopecia (AGA) and factors related to metabolic syndrome (MetS), which is known to increase the risk of type 2 diabetes mellitus and cardiovascular disease. However, the results of these studies have been inconsistent and most of them are based on prevalent survey rather than repeated surveys that preclude one from elucidating multi-step progression of AGA and also pinpointing whether the role of MetS plays in onset or progressive stage of the natural history of AGA. Furthermore, for a community-based study on these issues, the misclassifications of AGA status are usually an important concern. Therefore, studies on AGA progression and its association with MetS after considering misclassifications with an appropriate statistical method are worthy of being investigated. Objective: (1) To elucidate if there is an association between MetS and AGA after adjustment for potential confounders. (2) To estimate the progression rates of AGA and investigate its association with MetS with or without considering misclassification of AGA status. Methods: Population-based cross-sectional surveys were conducted in a Taiwanese community. Norwood and Ludwig classifications were used to assess the degree of hair loss in men and women. Information on components of MetS along with other possible risk factors was collected. (1) A total of 740 men aged 40 to 91 years participated in the survey between April and June 2005. The data were used to analyze the association between AGA and MetS. A logistic regression model was employed to assess the associations between MetS or each possible risk factor and the risk of moderate or severe AGA. (2) A total of 4,633 women and 2,362 men aged 30 to 95 years participated in the survey in 2005 and a total of 25,118 women and 16,884 men aged 30 to 102 years participated in the survey in 2010. A total of 899 women and 584 men aged 36 to 94 years participated in both surveys and they are used to estimate the incidence rates of AGA. Then, all of these data were utilized in estimation of AGA progression and its association with MetS. A multi-step Markov model was utilized for analysis. A Bayesian approach with Markov Chain Monte Carlo (MCMC) method for estimation of these parameters with correction for misclassifications was also used. Results: (1) A statistically significant association was found between AGA and the presence of the MetS (Odds ratio (OR) = 1.67, 95% CI: 1.01, 2.74) as well as between AGA and the number of fulfilled MetS components (OR= 1.21, 95% CI: 1.03, 1.42) after controlling for age, family history of AGA, and smoking status. Among MetS components, high-density lipoprotein (HDL) (OR= 2.36, 95% CI: 1.41, 3.95, p= 0.001) was revealed as the most important factor associated with AGA. (2) After 5-year follow-up, 89/745 (12.0%) women and 58/369 (15.7%) men developed AGA which leads to the incidence rates of 2.4% and 3.1% per person-year in women and men, respectively. In AGA progression, MetS was significantly associated with progression of AGA in the second-step transition (Hazard ratio (HR) =1.16, 95%CI: 1.04, 1.30) but lacking of statistically significant association with the first-step transition (HR=1.03, 95%CI: 0.98, 1.08) after adjusting for age, sex, and family history. Some individual components of MetS were found significantly associated with progression of AGA. These factors included lower serum HDL level in the first-step transition from normal to mild or moderate AGA (HR=1.07, 95%CI: 1.01, 1.13), fasting glucose or DM in both transitions (HR=1.06, 95%CI: 1.01, 1.11; HR=1.20, 95%CI: 1.08, 1.33), and hypertension in both transitions (HR=1.04, 95%CI: 1.01, 1.08; HR=1.16, 95%CI: 1.06, 1.26). These estimates are consistent after considering misclassifications which revealed estimates away from the null and larger standard errors. Conclusions: A statistically significant association was found between AGA and the presence of the MetS as well as between AGA and the number of fulfilled MetS components. With regard to progression of AGA, MetS was associated with the transition rate from mild or moderate to severe state of AGA which was revealed in both with or without considerations of misclassifications of AGA status. These results demonstrated a significant association between MetS and AGA. Identification of the MetS in moderate or severe AGA patients might be necessary for early recognition that would lead to early intervention to reduce the risk or complications of cardiovascular disease and type 2 diabetes mellitus later in life. In this study, we could also evaluate the extents of misclassifications for AGA and obtain the estimates for associations between MetS and AGA progressions after correcting for these measurements errors. These estimates could be used for predictions of transition probabilities which are useful for risk stratifications in population-based intervention studies on AGA. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/42253 |
全文授權: | 有償授權 |
顯示於系所單位: | 流行病學與預防醫學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-100-1.pdf 目前未授權公開取用 | 1.26 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。