比較兩種潛伏結核感染治療期程與都治介入方案之完治率─以臺北市為例

潛伏結核感染治療為根除結核病的重要策略。本研究為比較潛伏結核感染治療中，3個月Isoniazid與Rifapentine短程治療處方（簡稱3HP）及9個月Isoniazid治療處方（簡稱9H）之完治率，並探討行動線索：都治關懷，以及行動障礙：副作用發生情形對於完治率之影響。本研究採用衛生福利部傳染病監測系統疫情資料倉儲（BO）之結核病接觸者管理資料，經選取臺北市2016年1月至2017年9月之潛伏結核感染個案之治療紀錄進行分析。研究分析之個案共373人，97.9%加入都治方案，完治率達89.8%。其中317人接受3HP治療者，6.9%有發生副作用，89.3%完成治療；另56人接受9H治療處方者，僅1.8%發生副作用，92.8%完成治療，9H治療組之完治率高於3HP組，但沒有顯著不同。治療期間無發生副作用者的完治率，顯著高於有發生副作用者。9H治療組中，遠端都治個案的完治率95%，高於傳統都治個案之完治率89.3%，但無顯著差異。年齡為完成治療的預測因子，每增加1歲其完治的機率減少5%。建議持續推廣3HP短程的治療處方合併都治關懷治療潛伏結核感染者，並將遠端都治納入3HP都治關懷的服務選項，以強化高風險族群接受預防性治療的意願，使有效降低結核病發生之風險。

關鍵字

3HP ；健康信念模式；都治；結核病；潛伏結核感染治療

並列摘要

Treatment of latent tuberculosis infection (LTBI) is an important strategy toward TB elimination. The study was to evaluate the completion rate of 3 months of isoniazide and rifapentine regimen (3HP) and compared to 9 months of isoniazid regimen (9H). A retrospective data base study was employed to analyze the treatment of LTBI data from 2016 Jan to 2017 Sep in Taipei city from Business Objects (BO) system of Taiwan Centers for Disease Control (Taiwan CDC). Using Cox proportional hazards regression to examine the correlation of cues to action, DOPT, barriers of action, and side effect to the completion rate for latent tuberculosis infection treatment. The total sample cases included in this study was 374, while 97.9% of the cases joined DOPT and had a completion rate of 89.8%. The 3HP regimens included 317 cases. 6.9% of them had side effects and the completion rate was 89.3%. The 9HP regimens had 56 cases. 1.8% of them had side effects and the completion rate was 92.8%. The completion rate for cases without side effects were significantly higher than those with side effects. Cases in 9HP regimens using telecare DOPT had 95% of completion rate while the regular DOPT hold a 89.3% completion rate. Age is the major predictor of completion rate. Completion rate decreases 5% for every one year increase of age. The 3HP regimens for DOPT with telecare is recomanded to elevate the acceptence of preventive treatment for LTBI.