目標：檢視台灣地區道路密度與發生高血壓之相關性，及利用因果中介變項分析了解細懸浮微粒、交通噪音在關係中之角色。方法：研究利用全民健康保險2000年百萬承保抽樣歸人檔（Longitudinal Health Insurance Database [LHID] 2000）之世代追蹤研究資料，以確認2003年至2012年間罹患高血壓者的新病例。暴露資料則以交通部運輸研究所之路網數值圖2006第1.3版，透過地理資訊系統計算研究對象就醫鄉鎮之道路密度而細懸浮微粒資料取自2006至2012年空氣品質監測網之每小時PM_(2.5)自動監測資料，並以克力金模型推估至研究對象就醫地點所在之鄉鎮，做為受試者之長期暴露數值。最後再以Cox比例風險模型估計道路密度與發生高血壓之危險比及利用因果中介變項量化交通汙染所產生之細懸浮微粒所造成之高血壓效應（例：間接效應[Natural Indirect Effect, NIE]）。結果：研究對象所就醫之鄉鎮如有較高之縣道道路密度時（≥ 125.3 m/km^2），則其對之高血壓風險約略增加17%（危險比[Hazards Ratio, HR]=1.17, 95%信賴區間：1.14-1.19），而較高之鄉道道路密度（≥ 473.2 m/km^2）也發現相同的趨勢（危險比[Hazard Ratio, HR]=1.09, 95%信賴區間：1.07-1.10）。結論：本研究發現研究對象就醫鄉鎮之道路密度與發生高血壓有正向的相關性，此相關可能為細懸浮微粒所造成。

Objectives: To assess the association between roadway density and the incidence of hypertension and to evaluate the role of fine particulate matter (PM_(2.5)) on that association by using mediation analyses Methods: We included participants' health information retrieved from the Longitudinal Health Insurance Database 2000. We retrieved roadway density (e.g. county road) from the Institute of Transportation's Digital Road Network Map 2006. Participants' long-term exposure to traffic-related pollution was based on the roadway density of the township in which they resided. We applied Cox proportional hazards models to assess the association between the incidence of hypertension and road density. Casual mediation models were utilized to estimate the mediation effect of PM_(2.5) on the roadway density-hypertension association. Results: During a median follow-up time of 11.01 years, 108,733 cases of hypertension were identified. After we controlled for potential confounders, participants who resided in townships with a higher density of county roads (≥ 139 m/m^2) had a 17% higher risk for hypertension when compared to the reference group (Hazard Ratio [HR]=1.17; 95% Confidence Intervals [CIs]: 1.14-1.19), and similar results were also found for country roads (HR=1.09, 95% CIs: 1.07-1.10). Results were not materially different when we included only participants who lived in a metropolitan area (e.g. Taipei city). Furthermore, PM_(2.5) imposed a statistically significant mediating effect on the incidence of roadway density-associated hypertension for county roadway exposure (HR=1.01, 95% CIs: 1.00-1.02) and country roadway exposure (HR=1.001, 95% CIs: 1.00-1.02). Conclusions: This study provided evidence that roadway density was positively associated with the risk of hypertension in an Asian population. We also found that PM_(2.5) served as a mediator for this association.