臺灣北部愛滋病毒感染者之延遲診斷：2007至2018年之趨勢、危險因子與預後

Abstract

背景：延遲診斷的愛滋病毒感染者，容易發病為後天免疫缺乏症候群(AIDS)，不僅增加醫療上的負擔，亦增加死亡率，在公共衛生的角度上，更可能增加傳染他人的機會。但是在台灣針對愛滋病毒感染者的延遲診斷定義並無一定的共識，且缺乏對於延遲診斷的愛滋病毒感染者其危險因子與預後的暸解，因此本研究目的在瞭解愛滋病毒感染者延遲診斷的趨勢、相關的危險因子，並探討其預後。 方法：利用臺灣疾病管制署傳染病通報系統，研究對象為2007至2018年間臺灣北部的愛滋病毒感染者。將使用三種延遲診斷的定義進行區分：第一，六個月延遲診斷定義為「初次診斷為愛滋病毒感染者，於診斷後六個月內，其CD4小於350 cells/μL，或不論其CD4數值，發病為AIDS」；第二，三個月延遲診斷定義為「初次診斷為愛滋病毒感染者，於診斷後三個月內，不論其CD4數值，發病為AIDS」；第三、一個月延遲診斷定義為「初次診斷為愛滋病毒感染者，於診斷後一個月內，不論其CD4數值，進展為AIDS」。以logistic regression確認與延遲診斷有關的特性，再利用survival analysis與accelerated failure time (AFT)模型估計延遲診斷之愛滋病毒感染者存活情形。 結果：本研究共納入10042名愛滋病毒感染者，符合6個月延遲診斷者共5369名（53.5％）；符合3個月延遲診斷者共2722名（27.1％）；符合1個月延遲診斷者共2484名（24.7％）。且研究顯示利用三種延遲診斷定義進行逐年的比較，於2011年後皆有增加的趨勢。進一步瞭解與延遲診斷有關的特性則是男性、年齡較大、異性戀、教育程度較低及失業。此外，研究亦顯示延遲診斷的愛滋病毒感染者，其存活時間為非延遲診斷者的46%。 結論：本研究確認了愛滋病毒感染者與延遲診斷有關的社會人口特徵，應針對高風險族群，制定適宜的介入政策，使早期發現、早期診斷，降低愛滋病毒感染者延遲診斷的情形。

Background: Late diagnosis of infection with human immunodeficiency virus (HIV) has been associated with an increased risk of acquired immune deficiency syndrome (AIDS)-related illnesses, higher medical care costs, and increased morbidity and mortality. In public health, late diagnosis has also been associated with an increased risk of HIV transmission. However, there has been no consistent definition in reporting late diagnosis of HIV infection in Taiwan, and hence it remains little known about its factors and outcomes. This study aims to evaluate the trend, factors, and outcomes of late diagnosis of HIV infection in Northern Taiwan. Methods: Data on persons newly diagnosed with HIV aged ≥ 18 years in Northern Taiwan between 2007 and 2018 were extracted from the National Notifiable Diseases Surveillance System of Taiwan Center of Taiwan Centers for Disease Control (Taiwan CDC). Three types of late diagnosis were evaluated in this study: 6-month late diagnosis, 3-month late diagnosis, and 1-month late diagnosis. (1) 6-month late diagnosis was defined as persons reported with a CD4 count below 350 cells/mL or an AIDS-defining event, regardless of the CD4 cell count within 6 months after HIV diagnosis; (2) 3-month late diagnosis was defined as persons reported with an AIDS-defining event within 3 months after HIV diagnosis; (3) 1-month late diagnosis was defined as presenting with an AIDS-defining event within a month after HIV diagnosis. We used multivariable logistic regression to identify factors associated with late diagnosis; Kaplan-Meier estimates to evaluate the survival probability, and accelerated failure time model to compare the relative times between the group of late diagnosis and that without late diagnosis of HIV infections, in three different definitions. Results: Among 10042 HIV-positive persons included in this study, 5369 (53.5%) had the 6-month late diagnosis, 2,722 (27.1%) had the 3-month late diagnosis, and 2484 (24.7%) had the 1-month late diagnosis. The proportion of 6-month late diagnosis was increased from 48.8% in 2007 to 56.3% in 2018, 3-month late diagnosis was increased from 23.4% in 2007 to 32.5% in 2018, and 1-month late diagnosis was increased from 21.2% in 2007 to 30.8% in 2018. The correlates with all three definitions of late diagnosis were similar, with greater adjusted odds ratios for male sex (1.68 for the 6-month, 1.67 for the 3-month, and 1.55 for the 1-month), older age (e.g., for age  45 years versus 18-24 years being 3.42, 6.13, and 6.41, respectively), heterosexual (1.40, 1.56, and 1.47, respectively), unemployed or occupation unknown (1.30, 1.59, and 1.56, respectively), and year of HIV diagnosis later than 2009. Additionally, late diagnosis has been associated with an increased risk of all-cause deaths and AIDS-related deaths. The survival time of persons with late diagnosis as compared to that of persons without late diagnosis was 46% for the 6-month late diagnosis, and 22% for both the 3-month and 1-month late diagnosis. In addition to the late diagnosis, older age had a prominent effect on shortened survival time. Conclusions: There was increased slightly over time in the proportion of late diagnosis among HIV-positive persons in Northern Taiwan. Late diagnosis associated sociodemographic characteristics that were identified in this study offered interventional opportunities. Therefore, the need for strategies to earlier detection of HIV infection should be highlighted, especially among the risk population.