臺灣腦中風復發之統計預測模型:病例對照研究

A statistical model for predicting recurrence of stroke in Taiwan: case-control study

董志煒(Jhih-Wei Dong)

指導教授 ： 吳建華

中原大學/理學院/應用數學研究所/碩士(2021年)

https://doi.org/10.6840/cycu202100756

摘要

腦中風是造成失能和死亡的主要原因之一，而且中風復發更會加重失能程度和提升死亡風險。本研究主要的目標為利用多元羅吉斯迴歸建立中風復發的統計預測模型，另外，透過Cox比例危險迴歸模型來辨識出這些影響中風復發之危險因子。本研究是回溯性病例對照研究，研究從2000 ~ 2008年健保資料庫中，篩選出主診斷代碼為 430 ~ 437 並排除外傷者挑選受試者。排除2000 ~2001年有過中風經驗者，以2002 ~ 2008年間超過一次中風者為病例組，只有過一次中風者為對照組。若個案有過兩次中風以上，則計算兩次中風間隔時間；若個案只有一次中風，則計算第一次中風到最後追蹤日做為中風間隔時間，以中風的危險因子和中風間隔時間建立羅吉斯迴歸模型。結果模型的接收者操作特徵曲線下面積(Area Under the receiver operating characteristic Curve, AUC)為0.756。中風的危險因子依然在中風復發扮演要角: 高血壓(Hazard Ratio, \widehat{HR}=2.129)、心房顫動(\widehat{HR}=1.518)、糖尿病(\widehat{HR}=1.350)、冠狀心臟病(\widehat{HR}=1.303)以及男性(\widehat{HR}=1.226)。模型表現不夠理想可能是健保資料庫缺少了其他重要的危險因子的緣故。

關鍵字

中風 ； 中風復發 ； 復發預測 ； 健保資料庫

並列摘要

Cerebral Stroke is one of the main causes of disability and death. Moreover, the recurrence of stroke will aggravate the degree of disability and increase the risk of death. The main goal of this study is to use multivariate logistic regression to establish a statistical prediction model for stroke recurrence. In addition, to identify these risk factors that affect stroke recurrence through Cox proportional hazard regression model. This study is a retrospective case-control study. From the 2000-2008 National Health Insurance Research Database (NHIRD), the study selected the main diagnosis code as 430-437 and excluded the traumatic patients. Excluding those who had stroke experience from 2000 to 2001, those who had more than one stroke between 2002 and 2008 were the case group, and those who had only one stroke were the control group. If the case has had more than two strokes, the duration between the two strokes is calculated; if the case has only one stroke, the duration between the first stroke and the last tracking day is calculated as the stroke time. The logistic regression model was established based on the risk factors of stroke and the duration between strokes. As a result, the area under the receiver operating characteristic curve (AUC) of the model was 0.756. Stroke risk factors still play an important role in stroke recurrence: hypertension (Hazard Ratio, \widehat{HR}=2.129), atrial fibrillation (\widehat{HR}=1.518), diabetes (\widehat{HR}=1.350), coronary heart disease (\widehat{HR}=1.303), and men (\widehat{HR}=1.226). The unsatisfactory performance of the model may be due to the lack of other important risk factors in the NHIRD.

並列關鍵字

stroke ； recurrence of stroke ； NHIRD

參考文獻

1. World Health Organization. The top 10 causes of death. https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death

2. 衛生福利部統計處 (2017)。108年死因統計結果分析。衛生福利部統計處。 https://dep.mohw.gov.tw/DOS/cp-4927-54466-113.html

3. 衛生福利部 (2017)。中風預防人人有責! 90%的中風均與危險因子有關。衛生福利部。 https://www.mohw.gov.tw/fp-3567-37944-1.html

4. Vos, T., Lim, S. S., Abbafati, C., Abbas, K. M., Abbasi, M., Abbasifard, M., Abbasi-Kangevari, M., Abbastabar, H., Abd-Allah, F., Abdelalim, A., Abdollahi, M., Abdollahpour, I., Abolhassani, H., Aboyans, V., Abrams, E. M., Abreu, L. G., Abrigo, M. R. M., Abu-Raddad, L. J., Abushouk, A. I., … Murray, C. J. L. (2020). Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: A systematic analysis for the Global Burden of Disease Study 2019. The Lancet, 396(10258), 1204–1222. https://doi.org/10.1016/S0140-6736(20)30925-9

5. Institute for Health Metrics and Evaluation (IHME). GBD Compare Data Visualization. Seattle, WA: IHME, University of Washington, 2020. https://vizhub.healthdata.org/gbd-compare.