McEwan Philip, Ward Thomas, Chen Chien-Jen, Lee Mei-Hsuan, Yang Hwai-I, Kim Ray, L'Italien Gilbert, Yuan Yong
Centre for Health Economics, Swansea University, Wales, UK; Health Economics & Outcomes Research Ltd., Wales, UK.
Health Economics & Outcomes Research Ltd., Wales, UK.
Value Health Reg Issues. 2014 May;3:5-11. doi: 10.1016/j.vhri.2013.08.001. Epub 2013 Oct 30.
Hepatitis C virus (HCV) infection is the leading cause of liver disease, and Taiwan has among the highest prevalence of HCV infection in the general population in Northeast Asia, estimated at between 2% and 4%. The aim of this study was to estimate the number of patients living with chronic HCV infection in Taiwan and quantify the expected numbers in each of the five Metavir fibrosis stages.
We applied a back-projection approach, using observed hepatocellular carcinoma incidence between 1979 and 2008 and a smoothed Expectation-Maximization algorithm to maximize a Poisson likelihood to estimate the previous incidence of HCV infection. The algorithm was coded in Excel and combined with the MOdelling the NAtural histoRy and Cost-effectiveness of Hepatitis model (a hepatitis C natural history markov model) to predict the past and future numbers in each Metavir fibrosis stage.
Incident cases were predicted to have peaked in 1972 at 56,634 annually, with the prevalence peaking in 1986 at 763,737 infections and falling to 578,203 infections in 2012. It was estimated that in 2012, 127,795 (23.0%), 105,545 (19.0%), 81,211 (14.6%), 123,939 (22.3%), and 116,823 (21.1%) subjects were in fibrosis stages F0, F1, F2, F3, and F4, respectively.
Our study provides HCV infection prevalence estimates, stratified by Metavir fibrosis stage, in Taiwan for 2012. This has potential implications for budget planning, particularly with the availability of emerging therapies because fibrosis stage is predictive of both rapid and sustained virological response; therefore, planning expected treatment response in a given population could be enhanced with this additional information.
丙型肝炎病毒(HCV)感染是肝病的主要病因,台湾地区普通人群中HCV感染率在东北亚地区位居前列,估计在2%至4%之间。本研究旨在估算台湾地区慢性HCV感染患者的数量,并量化五个梅塔维(Metavir)纤维化阶段中每个阶段的预期患者数量。
我们采用了反向推算方法,利用1979年至2008年期间观察到的肝细胞癌发病率,并运用平滑期望最大化算法来最大化泊松似然,以估算既往HCV感染发病率。该算法用Excel编码,并与丙型肝炎自然史和成本效益模型(一种丙型肝炎自然史马尔可夫模型)相结合,以预测每个梅塔维纤维化阶段过去和未来的患者数量。
预计发病病例在1972年达到峰值,每年为56,634例,患病率在1986年达到峰值,为763,737例感染,到2012年降至578,203例感染。据估计,2012年分别有127,795例(23.0%)、105,545例(19.0%)、81,211例(14.6%)、123,939例(22.3%)和116,823例(21.1%)受试者处于纤维化阶段F0、F1、F2、F3和F4。
我们的研究提供了2012年台湾地区按梅塔维纤维化阶段分层的HCV感染患病率估计值。这对预算规划具有潜在意义,特别是考虑到新出现的治疗方法,因为纤维化阶段可预测快速和持续病毒学应答;因此,利用这些额外信息可更好地规划特定人群的预期治疗反应。