臺灣地區HIV與AIDS使用HAART治療之效益評估

Abstract

背景與目的 台灣在1984年出現首例愛滋病患，隨著醫療進步，雞尾酒療法（HAART）已經成為HIV感染者與AIDS病人主要治療方式。疾病管制署建立愛滋個案通報及管理系統，以公務預算與健保支應提供個案治療，另也致力於提供多種管道如匿名篩檢已找出潛在個案。國家層級的愛滋防制投入，應該針對其治療效益進行評估，本研究針對HIV與AIDS病患，以不同時期、年齡層、性別與傳染途徑納入分析，包含呈現台灣愛滋病疾病負擔以及在不同年齡層之流行病學現況、評估HIV感染以及進展至AIDS之病程、評估現行雞尾酒療法治療策略對HIV個案及愛滋病個案存活之介入效益、考量AIDS個案接受雞尾酒療法的延遲期後評估治療介入之效益。 研究方法 本研究採前瞻性世代研究， 疾管署愛滋資料庫所蒐集臺灣HIV感染及愛滋病現況去模擬一個1984年~2017年，含35000人的HIV感染前瞻性世代。資料包含病毒檢測、疾病進展、治療情況、存活率、人口統計資料、傳播方式等。 使用時間相依Cox比例風險回歸以評估HAART治療對愛滋病的影響，並考慮到人口統計學和時間順序對愛滋病存活率；另考慮潛伏期之於延遲治療對疾病進程影響，使用馬爾可夫模型進行分析。 研究結果 在模擬世代中，46.6%HIV感染個案有AIDS確診記錄，其中13.9%為死亡個案；HIV感染個案中有83.4%接受治療，其中半數是在確診AIDS之後，HIV/AIDS相關死因者佔4.38%。 25-34歲年齡層的HIV感染累積發生率最高、AIDS確診發生率也最高；整體HIV感染率約為每十萬人年有4.87例，愛滋病前之潛伏期約為10年。 以HIV感染者進行多變項分析結果而言，男性感染者的死亡風險較女性高26% (HR [95% CI]: 1.26 [1.08-1.47])，在模式中校正延遲治療效應及性別、年齡、感染時間、感染模式下同時評估治療介入的效益，治療介入後死亡風險減少63% (0.37 [0.33-0.42])；至於AIDS確診後的存活預後差異，男性感染者的死亡風險仍較女性高26% (HR [95% CI]: 1.26 [1.07-1.48])，在模式中校正延遲治療效應及性別、年齡、感染時間、感染模式下同時評估治療介入的效益，治療介入後死亡風險減少49% (0.51 [0.46-0.56])；無論HIV感染者與AIDS確診之後，高齡者的死亡風險都較高。 利用本研究資料，我們估計愛滋病在發病到接受治療的延遲期約為1.32年 (95%信賴區間：1.27-1.37)。進一步計HIV個案治療與否之HIV感染率以及進展至愛滋病之病程和預後，結果發現HIV感染率在無治療及有治療情況下分別約為每十萬人年4.31例(95%信賴區間：4.19-4.28)及8.20例(95%信賴區間：7.97-8.44)，愛滋病前之潛伏期在無治療及有治療個案分別約為12年(95%信賴區間：11.8-12.5年)及23.4年(95%信賴區間：23.2-23.8年)。因此，HAART治療對HIV個案而言可以減少約47%的愛滋病發病率(95%信賴區間：45-49%)，對愛滋病患而言可以減少約49%的死亡率(95%信賴區間：45-52%)。 結論 現行之防治對於HIV與AIDS族群效益顯著，然就本模擬研究中HIV與AIDS之疾病負擔年代趨勢顯示，HIV與AIDS之發生率仍分別約為10/100,000以及6/100,000。以年齡分層之分析結果顯示，年輕族群特別是15-34歲間其HIV感染風險大幅高於其他年齡層，AIDS 之罹病風險在25-45歲年齡層亦隨之上升。運用台灣實證資料評估此年輕族群在傳播模式以及地理區域之差異性以對於特殊高風險族群發展適合之感染防治措施，使現行以治療為主要模式之防治策略推進至以感染預防以及早期偵測為主之防治策略可作為台灣後續HIV與AIDS防治政策之發展目標。

Background and Objective： The provision of HAART (Highly Active Antiretroviral Therapy) to patients detected with AIDS and HIV have been implemented in Taiwan with an extended coverage by the infrastructure of health care system. Taiwan Center of Disease Control (TCDC) built the National HIV/AIDS Registry system (NHAR) for monitoring to monitor subjects with AIDS and HIV. In this study, we first explore the chronological trend of HIV and AIDS in Taiwan by periods and age groups to explore disease burden transformation through time. Based on progressive model from HIV to AIDS and the availability of HAART, we further evaluate the treatment efficacy by examining treatment delay and incubation time. Methods: We simulated a cohort of 35000 subjects with HIV derived from the NHAR 1984 to 2017 under the case management system of TCDC. The dataset contains information on the results of virus test, disease progression profile, treatment condition, survival rate, demographics, and transmission mode of the HIV/AIDS cases. The availability of such statistics is due to the physician’s necessity to report every identified case of the notifiable disease. Time-dependent Cox proportional hazards regression was applied to evaluate the effect of HAART on the AIDS subjects. The effect of demographics and chronological period on the HIV/AIDS survival rate, as well as a treatment delay estimated behind incubation period were also considered using the Markov model. Results: Among the simulated cohort, 46.6% HIV subjects progressed to AIDS, and 13.9% died from HIV/AIDS. 83.4% HIV subjects were treated with HAART, and among them half (50.2%) progress to AIDS. Epidemiological models for infectious diseases and the Markov process were adapted to assess the incidence and the disease progression pattern of HIV/AIDS. In view of the cumulative incidence trend, people under high risk of both HIV and AIDS are male aged from 25 to 34. The incubation period of the natural progression is around 10 years. Regarding the assessment for the effectiveness of HAART, both HIV survival rate and AIDS survival rate improve by period (decreased death risk manifested in death risk by period). Worse prognoses from the HIV subjects were observed in males (1.26 [1.08-1.47]) and cases without treatment (aHR:0.37 [0.33-0.42]). Furthermore, the AIDS subjects demonstrate the same pattern (male vs. female: 1.26 [1.07-1.48], treated vs. none: 0.51 [0.46-0.56]). Both HIV survival rate and AIDS survival rate were worse in older age groups. In this study, we estimate that the delay period from the onset of AIDS to treatment is about 1.32 years. Our further calculation reveals that the HIV infection rate（HIV to AIDS）without treatment to be 4.31 cases per 100,000 (95%CI：4.19-4.28), while the infection rate with treatment jumps to 8.20 cases per 100,000 (95% CI: 7.97-8.44). The incubation periods for the two groups before AIDS were about 12 years (95% CI： 11.8-12.5 years) and 23.4 years (95% CI：23.2-23.8 years). HAART treatment can reduce the incidence of AIDS by about 47% for HIV cases, and can reduce the mortality rate by about 49% for AIDS patients. Conclusions: The current prevention and treatment significantly benefit the HIV and AIDS groups. However, the chronological trend of the disease burden of HIV and AIDS in this simulation study shows that the incidence of HIV and AIDS is still respectively approximately 10/100,000 and 6/100,000. The analysis of age stratification presents the risk of HIV infection among young people, as the infection rate between 15 and 34 years old is significantly higher than the other age groups. The risk of AIDS from 25 to 45 also jumps out among all age groups. This study uses the Taiwanese empirical data to evaluate the differences in transmission patterns and geographic regions of the young people. It may also contribute in developing a suitable infection prevention method and control measures for the special high-risk groups, and in improving Taiwan’s current treatment-based prevention and treatment strategies.