Benefits and Effectiveness of Pneumococcal Immunization Policies for Young Children in Taiwan

背景：肺炎鏈球菌常引起幼兒及老人的嚴重疾病，尤其是侵襲性肺炎鏈球菌感染症(invasive pneumococcal disease, IPD)，患者預後常留下後遺症，甚至死亡。我國自2009年起，依風險訂定優先順序，分階段實施幼兒結合型肺炎鏈球菌疫苗 (pneumococcal conjuage vaccine, PCV) 接種計畫，效益亟待評估。 目的：一、評估我國5歲以下幼兒接種PCV之接種率及效果；二、對2-5歲幼兒接種一劑PCV13之效果；三、評估幼兒同時接種PCV及流感疫苗是否較接種單一疫苗有加成效益。 方法：一、利用疾管署的法定傳染病通報資料以及健保申報資料，分析全國5歲以下幼兒之PCV接種率、IPD發生率及健保相關醫療利用，採長期趨勢觀察。二、選取2011-2012年，針對2-5歲接受一劑PCV接種與完全未接種任何PCV疫苗的個案，採傾向分數配對法 (propensity score matching)以提高兩組間可比較性；串連健保資料庫，並利用負二項及廣義線性模式gamma迴歸模型分析其2013年與肺炎鏈球菌感染相關之醫療利用，以瞭解接種效果。三、選取2009-2013共4個流感季期間，針對5歲以下完成當季流感疫苗、4劑PCV、兩者均接種或均未接種共四組個案，串連資料庫，分析組間流感季中全死因死亡及肺炎鏈球菌感染相關醫療利用之差異，並進行成本效果分析。 結果：一、2012年起我國2-5歲幼兒IPD病例數顯著下降 (p<.0001)；2013年起小於2歲 (p<.01)及5-64歲 (p<.05)族群亦呈顯著下降趨勢；2012及2013年全國2-5歲幼兒相關健保費用分別較前8年平均減少17.5%及30.3%。二、2-5歲接種一劑PCV者因IPD門診的機率為未接種者之67%，因IPD住院的機率為未接種者之30% (p<.001)、因肺炎住院之機率為未接種之69% (p<.001)。接種疫苗者因IPD住院所花費的醫療費用為未接種者之68%、因肺炎住院所花費的醫療費用為未接種者之83% (p<.001)。三、各施打疫苗組均能減少肺炎及流感、呼吸道疾病及急性中耳炎之住院風險；施打PCV的兩組於流感季之全死因死亡風險僅為未施打組之18%及21%；惟同時施打兩種疫苗對降低住院及死亡風險並無加成效果。三個疫苗接種策略每拯救一個人年需要1.74到1.91個人均國內生產毛額，均具成本效果。以增量成本效果比 (ICER)來看，接種PCV雖較流感疫苗成本為高，但每增加拯救一個人年的費用仍具成本效果。 結論：我國實施山地離島、2-5歲及1歲接種PCV政策已顯著減少目標族群IPD發生率；2-5歲幼兒接種一劑PCV13能顯著減少因IPD門診及住院之風險，並減少相關醫療費用。接種PCV或PCV和流感疫苗可顯著減少流感季中8成死亡風險，且符合成本效果。故PCV單價雖高，但確為具成本效果之幼兒接種計畫，值得各國政府引進推廣。Background: Invasive pneumococcal disease (IPD) is one of the leading causes for hospitalization and deaths among children and the elderly. Pneumococcal conjugate vaccine (PCV) has been introduced by Taiwan government for different target groups of children under 5 years of age since 2009. The effectiveness of the programs has not yet been evaluated. Objectives: 1. To evaluate the PCV coverage and effectiveness for children under 5 years of age; 2. To assess the effectiveness of one dose regimen of PCV for 2-5 years of age; 3. To evaluate the synergy effects for co-administration of PCV and influenza vaccine. Methods: 1. We analyzed long term trend of PCV coverage, IPD prevalence and medical utilization among children under 5 years of age in Taiwan by using CDC registration data and National Health Insurance claim data. 2. For children aged 2-5 years old in 2011-2012, we evaluated the one-dose regimen by comparing medical utilization between vaccinated and non-vaccinated groups. Propensity score matching (1:2) was used to increase the comparability of the two groups; the statistics methods used were generalized linear model with negative binomial regression and logarithmic link and gamma distribution. 3. For children under 5 years of age during 4 influenza seasons from 2009-2013, we classified the subjects into 4 groups: influenza vaccinated, 4-dose PCV vaccinated, both, or none, and compared the all-cause death and medical utilization. In addition, a cost-effectiveness analysis (CEA) was also conducted in this study. Results: 1. The number of IPD cases among children 2-5 years of age have decreased significantly since 2012 (p<.0001); the number of cases under 2 (p<.01) and 5-64 (p<.05) years old have also decreased since 2013. Compared with previous 8 years, pneumococcal infection related costs claimed for National Health Insurance among children 2-5 years of age in 2012 and 2013 decreased 17.5% and 30.3%, respectively. 2. OPD visits related to IPD in vaccinated group is 33% less than that of unvaccinated group (p<.001). Hospitalization related to IPD and pneumonia in vaccinated group is 70% and 31% less than that of unvaccinated group, respectively (p<.001); the cost related to IPD and pneumonia in vaccinated group is 32% and 17% (p<.001) less than unvaccinated group, respectively. 3. Compared with none-vaccinated groups, the three vaccinated groups were all with lower risks for hospitalization caused by pneumonia & flu, respiratory disease and acute otitis media (AOM). The risks of all-cause death of PCV and PCV+FLU groups during flu season were only 18% and 21% of the none vaccinated group, respectively (p<.0001). Nevertheless, there is no synergy effect for those receiving both PCV and flu vaccines. The three vaccination strategies were considered cost-effective with 1.74-1.91 Gross Domestic Product (GDP) per capita needed per life-year gained. The cost of PCV vaccination strategy was higher than influenza vaccination; yet it was still cost-effective with an incremental cost-effectiveness ratio (ICER) per life-year gained reached the WHO criterion. Conclusion: PCV vaccination programs for children were not only effective but also cost-effective; it’s worthy to be included in the national programs