Post-marketing safety surveillance for inactivated and live-attenuated Japanese encephalitis vaccines in China, 2008-2013

1 Pos -ma ke ing sa e y su eillance o inac i a ed and li e-a enua ed Japanese encephali is accines in China, 2008–2013 Wu Wendia,b,1, Liu Daweia,1, Li Kelia*, Pekka Nuo ib, Hanna M. Nohynekc, Xu Dishaa, Ye Jiakaia, Zheng Jingshana, Wang Huaqinga,*. a Na ional Immuniza ion P og amme, Chinese Cen e o Disease Con ol and P e en ion b School o Heal h Sciences, FIN-33014 UNIVERSITY OF TAMPERE, FINLAND c Na ional Ins i u e o Heal h and Wel a e THL Helsinki Finland 1 Wu Wendi and Liu Dawei con ibu ed equally o his wo k. * Co esponding au ho s:Wang Huaqing and Li Keli. Tel:+86 10 63171892(Wang)/83166393- 1(Li), E-mail add ess: [email protected](Wang)/[email protected](Li) Abs ac In oduc ion: Two ypes o Japanese encephali is (JE) accines, inac i a ed JE accine (JE-I) and li e- a enua ed JE accine (JE-L), a e a ailable and used in China. In pa icula , one JE-L, p oduced by a domes ic manu ac u e in China, was p equali ied by WHO in 2013. We assessed he sa e y o JE accines in China du ing 2008-2013 using he Chinese Na ional Ad e se E en s Following Immuniza ion In o ma ion Sys em (CNAEFIS) da a. Me hods: We e ie ed AEFI epo ing da a abou JE accines om CNAEFIS, 2008-2013, examined demog aphic cha ac e is ics o AEFI cases, and used adminis a i e da a on accine doses as denomina o o calcula e and compa e c ude epo ing a es. We also used disp opo ionali y epo ing analysis be ween JE-I and JE-L o assess po en ial sa e y signals. Resul s: A o al o 34,879 AEFIs ela ed wi h JE-I and JE-L we e epo ed, wi h a a io o male o emale as 1.3:1; 361 (1.0%) cases we e classi ied as se ious. JE accines we e adminis e ed concu en ly wi h one o mo e o he accines in 13,592 (39.0%) o cases. The o e all AEFI epo ing a es we e 214.4 pe million accina ion doses o JE-L and 176.9 o JE-I ( a e a io [RR]: 1.2, 95% con idence in e al [CI]: 1.1-1.3) in 2010-2013. Feb ile con ulsions (FC) ollowing JE-I was ound as a signal o disp opo iona e epo ing (SDR). Howe e , he e was no signi ican di e ence be ween he epo ing a es o FC o JE-I and JE-L (0.3 pe million accina ion doses o JE-L, 0.4 o JE-I, p=0.05). Conclusions: While ou analysis did no ind appa en sa e y conce n o JE accines in China, u he s udy should conside conside JE-I accines and eb ile con ulsion, and aking mo e sensi i e me hods o de ec signals. Key wo ds ： Vaccine sa e y, ad e se e en s ollowing immuniza ion, eb ile con ulsion, Japanese encephali is accine, disp opo ionali y analysis, su eillance sys em, China 2 In oduc ion Japanese encephali is (JE) is a mosqui o-bo ne acu e i al in ec ion o he cen al ne ous sys em caused by a la i i us [1]. JE is he mos impo an cause o accine-p e en able i al encephali is in nea ly all Asian coun ies, whe he empe a e, sub opical, o opical, and has expanded in o new a eas h ough he impo a ion o in ec ed-mosqui o ec o s. Cu en ly, an es ima ed 3 billion people li ing in 24 coun ies, mainly in he Sou h-Eas Asia and Wes e n Paci ic Regions a e conside ed a isk o JE [2]. The inac i a ed JE accine (JE-I) was de eloped in China and has been used since he 1970s and li e-a enua ed accine (JE-L) was in he beginning o he 1990s [1]. Since 2007, JE accines we e included in o he Expanded P og am on Immuniza ion (EPI) in he mainland o China [1]. Wi h he decline o numbe o JE disease in China, he public became mo e conce n abou he ad e se e en s ollowing JE accina ion cu en ly. The sa e y o JE accines manu ac u ed in China and ab oad was e alua ed in p e ious clinical and pos -ma ke ing s udies [3–7]. The accine sa e y e iew o JE accines by Wo ld Heal h o ganiza ion (WHO) we e ound o ha e accep able sa e y p o iles, and da a om mul iple s udies (including mul icen e andomized con olled ail and andomized ials) had shown he same conclusion [2,8-9]. Howe e , The JE accine used in China we e mainly p oduced by domes ic manu ac u e s, and a JE-L p oduc was p equali ied by WHO in 2013, which was he i s Chinese-p oduced accine o be p equali ied by WHO. Limi ed da a a e a ailable on he sa e y o JE a e i s inclusion in o he Chinese EPI and consequen ly i s la ge-scale use. Concu en ly wi h he inclusion o JE accines in o EPI, he Chinese na ional ad e se e en ollowing immuniza ion (AEFI) in o ma ion sys em (CNAEFIS), a passi e pos -ma ke ing accine sa e y su eillance sys em, was also expanded o co e he en i e coun y a e a 3-yea pilo s udy [10,11]. Wi h 6 yea s o in o ma ion collec ion, he e we e aluable AEFI da a in he CNAEFIS which could p o ide some e idence o e alua ion o accine sa e y in China. Also, using he passi e su eillance da a, he disp opo ionali y analysis, which was i s used in signal de ec ion o d ug sa e y, could also be necessa y o gene a e signals, especially when he causali y o he speci ic e en s and accina ion has no been well known. We conduc ed a s udy o analyze AEFI da a o bo h JE-I and JE-L accines in CNAEFIS om 2008 o 2013 o unde s and and compa e he accine sa e y p o iles o hese 2 accines. 3 Me hods Vaccina ion schedules o JE accines in China [10] In he mainland o China, bo h JE-I and JE-L ha e been included in Na ional Immuniza ion P og am (NIP) accina ions since 2007. Fo JE-L, a 2-dose schedule is used: a 8 and 24 mon hs o age, wi h a leas 3 mon hs’ in e al. Fo JE-I, a 4-dose schedule is used (2 doses a 8 mon hs wi h a leas 7–10 day in e als, and subsequen doses a 2 and 6 yea s o age). Vaccina ion doses o JE accines All he accines should be used in Vaccina ion clinics, which was app o ed by local go e nmen and supe ised by local Cen e o diseases con ol and p e en ion (CDC). Vaccina ion doc o s o nu ses collec ed in o ma ion o accina ion doses and epo ed o he coun y CDCs mon hly. The coun y CDCs epo he da a o municipal CDCs who epo o p o incial CDCs. China CDC collec s adminis e ed accina ion da a om all p o incial CDCs [3, 13]. Be o e 2010, only doses o accines p o ided by he go e nmen o ee we e epo ed. The e o e, comple e da a on all JE accina ion doses gi en du ing 2008–2009 a e no a ailable. Since 2010, howe e , he accina ion in o ma ion sys em collec ed all accina ion doses adminis e ed in he accina ion clinics, enabling he analysis o he epo ing a es. As he denomina o in o ma ion only included he numbe o accine doses, wi hou in o ma ion on age o sex, he a es o AEFI o speci ic popula ion g oups by age o gende could no be calcula ed. CNAEFIS The online CNAEFIS is adminis e ed by he Chinese Cen e o Disease Con ol and P e en ion (CCDC). A e wo pilo s udies, i was expanded o co e all 31 p o inces in he mainland o China in 2008 [10]. The pilo s udies included a passi e su eillance sys em o AEFI in 10 p o inces and an enhanced AEFI su eillance sys em in 5 coun ies. CNAEFIS is ope a ed in acco dance wi h China’s na ional AEFI guidelines [12]. Acco ding o his guidance, CNAEFIS became he o icial AEFI in o ma ion sys em and was o be owned and main ained by China CDC [10]. An AEFI case is de ined as a eac ion o an e en occu ing a e accine adminis a ion ha is suspec ed o be ela ed o he accina ion. AEFI su eillance and epo ing co e s all accines ma ke ed in he mainland o China [10]. Repo ing and In es iga ion [10-13] Heal hca e acili ies, accina ion clinics, Cen e s o Disease Con ol and P e en ion (CDCs) a all 4 adminis a i e le els, ad e se d ug eac ion moni o ing agencies (ADRs), and accine manu ac u e s a e equi ed by law o epo suspec ed AEFIs. The public o he gua dian (pa en s) can no i y any o he abo e au ho ized epo e s o epo an AEFI. Cases a e in es iga ed by local, coun y-le el CDCs, which a e esponsible o comple ing AEFI Case Repo ing Ca ds and submi ing he da a o online CNAEFIS. Once he in o ma ion is en e ed, i can be iewed by all adminis a i e le els o CDCs and ADRs. Based on he add ess, name and bi hday o he child, accines, and accina ion da es, duplica e epo s a e iden i ied and po en ial mul iple epo s a e combined in o one case. 4 In es iga ion is equi ed o all AEFIs, excep common ad e se eac ions wi h a clea diagnosis (e.g., e e ; edness, swelling, and indu a ion on he injec ion si e). Fo dea hs, se ious AEFIs, AEFI clus e s, and AEFIs o signi ican public conce n ha a e suspec ed o be ela ed o accina ion, p e ec u al o p o incial CDCs mus immedia ely o ganize an AEFI expe panel o in es iga ion upon ecei ing CNAEFIS epo s. Se ious and non-se ious AEFIs [13] Se ious AEFI is de ined as an e en ha is causing a po en ial isk o he heal h/li e o a ecipien leading o p olonged hospi aliza ion, disabili y/incapaci y, congeni al abno mali ies/ bi h de ec s o dea h. In CNAEFIS, i include, bu a e no limi ed o, alle gic shock, alle gic la yngeal edema, alle gic pu pu a, h ombocy openic pu pu a, localized alle gic nec o ic eac ion (A hus eac ion), eb ile con ulsion, epilepsy, b achial neu i is, polyneu i is, Guillain–Ba e synd ome, encephalopa hy, encephali is and meningi is, syncope, oxic shock synd ome, and sys emic pu ulen in ec ion. Da a analysis We analyzed AEFI epo s submi ed du ing 2008–2013, o subjec s accina ed wi h JE-L o JE-I. In CNAEFIS, a maximum 3 suspec ed accines can be epo ed a he same ime in a single epo . JE accines lis ed as he i s , second, o hi d suspec ed accine we e all included. When mo e han one symp om was epo ed o a case, only he main symp om o he mos se ious diagnosis was eco ded in CNAEFIS. The age and sex dis ibu ion and clinical diagnoses we e desc ibed, and c ude AEFI epo ing a es pe million doses gi en we e calcula ed. Since he e is no in o ma ion on whe he he accines we e NIP accine o no in AEFI cases, he incidence o NIP accine o olun a y accina ion could no be es ima ed du ing he s udy yea . We used disp opo ionali y analysis o da a mining algo i hms o compa e he equency o epo s o JE-L and JE-I o de ec any signal o disp opo iona e epo ing (SDR) [14]. Disp opo ionali y analysis iden i ies AEFIs ha we e mo e equen han expec ed and elies on he p inciple ha when a SDR is iden i ied o a speci ic accine, his e en (o diagnosis) is epo ed ela i ely mo e equen ly in associa ion wi h his speci ic accine han all he o he accine in he da abase. Th ee disp opo ionali y analysis me hods we e applied: he p opo ional epo ing a io (PRR) [15–16], Bayesian con idence p opaga ion neu al ne wo k (BCPNN) [17], and empi ical Bayesian (EB) da a mining [18–20]. Disp opo ionali y analysis we e based on a 2*2 con ingency able simila o a case-con ol s udy o coho s udy [21]. Calcula ions we e pe o med using R ( e sion i386 3.2.3), and he PhViD package we e used in analysis. Since disp opo ionali y analysis equi ed “ accine and diagnosis” as a pai , cases wi hou con i med clinical diagnosis we e excluded. Fo he cases in which diagnosis included common mino ad e se eac ions, wi h a mix o symp oms such as e e , local edness, local swelling, and o he mino local o sys emic symp oms, “common eac ion” was used as a single diagnosis. Cases who had ecei ed concu en accines we e excluded. 5 Resul s AEFI ollowing JE-L and JE-I A o al o 34,879 AEFI cases associa ed wi h JE accines we e collec ed by CNAEFIS, 2008- 2013; 95.2% (33,186) cases we e ela ed o JE-L. JE accines we e adminis e ed concu en ly wi h one o mo e o he accines in 13,592 (39.0%) o cases (39.9% o JE-L and 19.9% o JE-I, espec i ely). Bo h o JE-L and JE-I, he mos common concu en ly adminis e ed accine was measles- con aining accines, wi h a p opo ion o 24.8% o JE-L (8226 cases in 33186 JE-L- ela ed cases) and 17.5% o JE-I (297 cases in 1693 JE-I- ela ed cases). JE-L was lis ed as he i s suspec ed accine in 23,627 (71.2% o he JE-L-associa ed AEFI cases). JE-I was he i s suspec ed accine in 1357 (80.2% o JE-I-associa ed cases) (p<0.05). The e we e mo e cases in males han in emales, wi h a sex a io o 1.3:1. Mo e cases occu ed in ≤1 yea s o age, wi h 66.4% o JE-L and 60.8% o JE-I (Table 1). O all 34,879 AEFI cases, 361 (1.0%) AEFI cases we e de ined as se ious. The e we e 146.7 million accina ion doses collec ed o JE accines om 2010-2013, in which 95.1% (139.5 million doses) was JE-L. Since bo h JE-L and JE-I could be used as NIP accines and olun a y accines, among all JE accina ion doses, 91.5% (134.3 million doses) we e used as NIP accines, including 133.1 million o JE-L and 1.2 million o JE-I. Using he doses adminis e ed om 2010 o 2013 as denomina o s, he o e all epo ing a es o AEFIs pe million we e 214.4 o JE-L and 176.9 o JE-I (RR: 1.2, 95% CI: 1.1–1.3). he annual epo ing a es inc eased subs an ially om 2010 o 2013 (Table 2). Du ing 2010–2013, 271 se ious AEFIs we e epo ed. The o e all epo ing a es o se ious AEFIs we e 1.8 pe million doses o JE-L and 2.8 pe million doses o JE-I (RR: 0.7, 95% CI: 0.4–1.0). Clinical diagnosis o AEFIs O he 29,831 non-se ious AEFIs, (86.4%) we e diagnosed as common and mino ad e se eac ions, such as e e , local edness, and swelling. Among se ious AEFIs, he mos equen ly epo ed clinical diagnosis we e eb ile con ulsion (132, 36.6%), h ombocy openic pu pu a (39, 10.8%), encephali is and meningi is (29, 8.0%), Henoch- Schönlein pu pu a (28, 7.8%), and anaphylac ic shock (25, 6.9%). (Table 3) 22 dea h cases we e epo ed du ing he s udy pe iod, only 4 cases we e classi ied as ela ed o accina ion due o anaphylac ic shock. Disp opo ionali y analysis A o al o 20,988 AEFIs wi h comple e in o ma ion on accine and diagnosis we e included in he disp opo ionali y analysis. All h ee me hods, PRR, EB, and BCPNN, sugges ed JE-I and eb ile con ulsion as he suspec ed SDRs (Table 3). Fo JE-L, he e was no diagnosis wi h disp opo ionally highe epo ing. Based on he esul s o Table 4, using he adminis e ed doses om 2010–2013 as he denomina o , he es ima ed epo ing a es o eb ile con ulsion a e JE-L (as he only accine suspec ed) and JE-I (as he only accine suspec ed) we e calcula ed: 0.3 pe million doses o JE-L and 0.4 pe million doses o JE-I (p=0.5). 6 Discussion We e iewed epo ed AEFIs a e JE-L and JE-I in CNAEFIS by 3 le els o analysis: a desc ip ion o he cha ac e is ics o AEFIs, disp opo iona e epo ing be ween JE-L and JE-I and es ima ed epo ing a es. The se ious AEFIs only accoun ed o 1% o all epo ed AEFIs, and he e was also no s a is ical di e ence be ween JE-L and JE-I in he epo ing a es o se ious AEFI in all 2010– 2013.Howe e in 2012, he epo ing a es o se ious AEFI ollowing JE-L we e lowe han JE-I. The inc easing end in he epo ing a es o AEFI ollowing he immuniza ion o JE-L and JE-I may be ela ed o he imp o ed epo ing o CNAEFIS du ing 2008 o 2013. The inc eased ends we e simila o he epo ing a es o all AEFIs du ing he same pe iod; he epo ing a es o all AEFIs in CNAEFIS o China con inuously inc eased om 2010 o 2013 [22–25]. In p e ious CNAEFIS da a analysis, only JE accines we e included in he s udy [20–23]. Howe e , in ou s udy, a leas one hi d o AEFIs had o he concu en ly adminis e ed accines. In addi ion, in he ≤1-yea g oup, which was he a ge g oup o EPI, an in AEFI o JE-L, mo e han hal he cases we e ela ed o JE-L in combina ion wi h o he accines. In hese si ua ions, i is di icul o de e mine he esponsible accine o some ad e se e en s, such as alle gic eac ions. Since concu en accina ion could no be a oided, we included all AEFIs ela ed o JE in he CNAEFIS da abase. The gende dis ibu ion o JE-L and JE-I AEFIs cases was simila o all o he accines [22-24], as he e we e mo e male han emale cases. Mos epo ed cases we e ≤1 yea old, consis en wi h he China Immuniza ion schedule o EPI accines. Mos epo ed AEFIs we e ela i ely mild and sel - limi ing, wi h a diagnosis o common and mino ad e se eac ions, including e e , local edness and swelling, and local indu a ion. Besides he mild and sel -limi ing ad e se eac ions, nea ly 97% o cases eco e ed, and only 22 dea h cases we e epo ed du ing he s udy pe iod. Cu en ly, he e a e se e al JE accines used in China. Abou 8 domes ic manu ac u e s in China p oduce JE-L and JE-I. In he mainland o China, since 1968, JE-I accines (Hams e kidney cell inac i a ed accine) we e manu ac u ed domes ically. The mass accina ion pe iod o JE was a e 1989, he yea JE-L was de eloped and manu ac u ed by Chengdu ins i u e o biological p oduc s company, and un il 2001, he yield o JE-L was mo e han 200 million doses in all. Following he posi i e assessmen by he WHO o China’s accine Na ional Regula o y Au ho i y in 2011 [25] and wi h he de elopmen o a accine manu ac o y in China [26], JE-L was he i s accine in China o be p equali ied by he WHO [27]. Di e en JE accines ha e been used in o he coun ies. Fo example, he Uni ed S a es (US) used an inac i a ed mouse b ain-de i ed JE accine (JE-I) o a ele s (mos o hem we e adul s) om 1992 o 2009. Japan used JE-I o child en. Takahashi [4] e iewed he pos -ma ke ing su eillance o JE-I om Japan and he US. The o al ad e se e en s a e was 2.8 pe 100,000 doses in Japan and 15.0 pe 100,000 doses in he US. JE-L was widely used in coun ies in he Wes e n Paci ic egion and Asia (China, Ko ea, and India), bu e y limi ed da a a e a ailable on he ad e se e ec s o JE-L in hese coun ies. An analysis o he pos -ma ke ing su eillance o JE-L om 2009–2012 by he Chinese Na ional Cen e o Ad e se D ug Reac ion (ADR) Moni o ing [2] showed ha in 6024 AEFIs, only 70 we e conside ed se e e. The Global Ad iso y Commi ee on Vaccine Sa e y (GACVS) o he WHO e iewed hese da a and no ed ha al hough he e was no e idence o a sa e y signal, he numbe o e en s eco ded in he AEFI epo ing sys em was low, gi en ha >70 million doses o he accine ha e been adminis e ed [25]. 7 One o he p ima y goals o AEFI passi e su eillance is o de ec accine sa e y signals and gene a e hypo heses o u he s udies. Using accina ion doses as a denomina o , AEFI incidence a es a e calcula ed. Th ough a compa ison wi h his o ic da a and published s udies, eme ging accine sa e y signals a e de ec ed. Howe e , since se ious AEFIs a e e y a e, and i is di icul o de e mine he backg ound incidence a e, in ecen decades, esea che s on pha maco igilance ha e applied da a mining me hods in he de ec ion o d ug ad e se e en s’ signals on accines [28]. One o he main analysis me hods is disp opo ionali y analysis. In he US, PRR [15-16], which was also used in he UK Yellow Ca d Scheme (YCS) ( he spon aneous epo ing sys em in he UK), and he Gamma Poisson Sh inkage model (GPS), which was also no ed as EB da a mining, has been used o sc een Vaccine Ad e se E en s Repo ing Sys em (VAERS) da a and gene a e signals, including signals om seasonal in luenza accines and om newly licensed accines [18-20]. The WHO Uppsala Moni o ing Cen e (UMC) has used a simila da a mining me hod, BCPNN, o de ec sa e y signals [17, 29]. These me hods we e no used as ou ine SDR sc eening in CNAEFIS, bu we applied he me hods o compa e he SDR be ween JE-L and JE-I as supplemen a y analysis o e alua e he sa e y o JE in China. Feb ile con ulsions a e he commones ype o seizu e in child en occu ing in 2–5% o all child en, 2–5% o young child en in No h Ame ica and Eu ope, and 6–9% in Japan [30-31]. I usually occu s be ween 3 mon hs and 5 yea s, wi h a peak incidence a 18 mon hs [32], which was also he a ge age o accina ions, especially o JE-L and JE-I. Gene ally, a leas 50% o child en who p esen wi h eb ile con ulsion will ha e no iden i ied isk ac o s [33]. As ele a ed body empe a u e is equen ly obse ed ollowing immuniza ion, and eb ile seizu es a e he mos common seizu e diso de in in an s and child en, hey a e he mos common ype o non-epilep ic seizu e obse ed ollowing immuniza ion [33]. The signal o eb ile con ulsion has been ound a e measles-mumps- ubella combined accine (MMR), diph he ia- e anus-pe ussis combined accine (DTP), and seasonal in luenza accine in some coun ies, such as he US and Aus alia [34–37]. Howe e , an inc eased isk o eb ile con ulsion a e JE has a ely been epo ed. In a andomized ial wi h 26 239 subjec s, wi hin 30 days ollow-up, he incidence p opo ion o seizu e was 0.1%, and e e las ing mo e han 3 days was 2.7% in he g oup accina ed wi h JE-L. The e was no s a is ically signi ican di e ence be ween accina ed and un accina ed g oups o seizu e o e e las ing mo e han 3 days [38]. In ou s udy, using he 3 da a-mining me hods, an SDR: JE-I and eb ile con ulsion was de ec ed compa e wi h JE-L. This indica ed ha he e was a disp opo ionali y in epo s o eb ile con ulsion a e JE-I compa ed o JE-L. Howe e , when compa ing he o e all es ima ed epo ing a es o eb ile con ulsion a e bo h accines, he e was no signi ican di e ence. In addi ion, eb ile seizu es a e JE we e less common han a e measles-con ained accines and a icella accines [36]. The e o e, he SDR o JE-I and eb ile con ulsion should be in e p e ed wi h cau ion. CNAEFIS, as a na ional passi e su eillance sys em, has i s s eng hs and limi a ions. E en i i se es as a use ul sou ce o accina ion sa e y e alua ion in China, he indings in CNAEFIS should be in e p e ed wi h cau ion. The na u al limi a ions o passi e su eillance include o e - and unde - epo ing, biased epo ing, and inconsis ency in he quali y and comple eness o epo s among o he s [37]. Meanwhile, he clinical diagnosis o AEFI cases a ied in p o inces o China, as we did no ha e a s anda d p ocedu e o de ini ion o AEFI diagnosis, and we did no use he s anda d de ini ions c ea ed by he B igh on Collabo a ion [39]. In ou analysis, since CNAEFIS we e case-based epo ing sys em, and accina ion doses we e collec ed based on summa ized ables om accina ion clinics, hese 2 da a esou ces we e no exac ma ched. The e we e no in o ma ion on whe he he accine used as NIP accine o olun a y accines in CNAEFIS, and no 8 gende and age in o ma ion on accina ion doses, all hose ela ed incidence could no be es ima ed using cu en da a. In conclusion, du ing he s udy pe iod o 2008–2013, he majo i y o AEFIs ollowing JE accines we e mino and common ad e se eac ions, and he e was no signi ican di e ence be ween he es ima ed epo ing a es o se ious AEFI ollowing JE-I and JE-L. Fo se ious AEFIs, mo e han 97% o cases eco e ed. We ecommend conside u he s udy o disce n e ec s o concu en accina ion wi h JE accines and ake mo e sensi i e me hods o de ec signals. Re e ences [1] Wang Huanyu, Li Yixing, Liang Xiao eng, e al. Japanese Encephali is in Mainland China. Jpn J In ec Dis 2009, 62:331-336. [2] Wo ld Heal h O ganiza ion. Japanese Encephali is Vaccines: WHO posi ion pape -Feb ua y 2015. WER 2015, 90(6): 69-88. [3] Liu Yu, Lin Hualiang, Zhu Qi, e al. Sa e y o Japanese encephali is li e a enua ed accina ion in pos -ma ke ing su eillance in Gyangdong, China, 2005-2012. Vaccine 2014, 32:1768-1773. [4] Hi oshi Takahashi, Vi ali Pool, Theodo e F Tsai, e al. Ad e se e en s a e Japanese encephali is accina ion: e iew o pos -ma ke ing su eillance da a om Japan and he Uni ed S a es. Vaccine 2000,18: 2963-2969. [5] Te suo Nakayama, Kazumasa Onoda. Vaccnie ad e se e en s epo ed in pos -ma ke ing s udy o he Ki asa o Ins i u e om 1994-2004. Vaccnie 2007, 25: 570-576. [6] H D No hdu , T Jelinek, A Ma schang, e al. Ad e se eac ions o Japanese encephali is accine in a ele s. Jou nal o In ec ion 1996, 32: 119-122. [7] Tom Solomon, Nguyen Minh Dung, Rachel Kneen, e al. Japanese encephali is. J Neu ol Neu osu g Psychia y 2000;68:405-415. [8] Chokephaibulki K, Si i ichayakul C, Thisyako n U, e al. Sa e y and immunogenici y o a single adminis a ion o li e-a enua ed Japanese encephali is accine in p e iously p imed 2- o 5-yea -olds and nai e 12- o 24-mon h-olds: mul icen e andomized con olled ial. Pedia In ec Dis J,2010;29(12):1111–1117. [9] Liu ZL, Hennessy S, S om BL, e al. Sho - e m sa e y o li e a enua ed Japanese encephali is accine (SA1414- 2): esul s o a andomized ial wi h 26,239 subjec s. J In ec Dis,1997;176(5):1366–1369. [10] Liu Dawei, Wu Wendi, Li Keli, e al. Su eillance o ad e se e en s ollowing immuniza ion in China: Pas ,p esen , and u u e. Vaccine 2015, 33: 4041–4046. [11] Liang Xiao eng, Li Li, Liu Dawei, e al. Sa e y o In luenza A (H1N1) Vaccine in Pos ma ke ing Su eillance in China. N Engl J Med 2011, 364: 638-47. [12] China Minis y o Heal h, Food and D ug adminis a ion o China. Na ional guideline o he su eillance o ad e se e en s ollowing immuniza ion. Beijing: China Minis y o Heal h, Food and D ug adminis a ion o China; 2010. [13] China Minis y o Heal h. Chinese S anda d P ocedu es o Vaccina ion. Beijing: China Minis y o Heal h; 2005. [14] Man ed Hauben, Je ey K A onson. De ining ‘Signal’ and i s sub ypes in Pha maco igilance based on a sys ema ic e iew o p e ious de ini ions. D ug sa e y 2009, 32(2): 99-110. 9 [15] E ans S.J.W., Walle P.C., Da is S. Use o P opo ional epo ing a ios (PRR) o signal gene a ion om spon aneous ad e se d ug eac ion epo s. Pha macoepidemiology and D ug Sa e y 2001; 10: 483-486. [16] Eu opean Medicines Agency. Eud a Vigilance Expe Wo king G oup. Guideline on he use o s a is ical signal de ec ion me hods in he Eud a Vigilance da a analysis sys em. (EMEA/106464/2006 e . 1) [Online]. A ailable om URL: h p://www.ema.eu opa.eu/docs/en_GB/documen _lib a y/Regula o y_and_p ocedu al_guideline/2009/11/WC5000114 34.pd [Accessed 2015.8.11] [17] And ew Ba e, Ma ie Lindquis , I Ralph Edwa ds, e al. A da a mining app oach o signal de ec ion and analysis. D ug sa e y 2002, 25(6): 393-397. [18] DuMouchel W. Bayesian da a mining in la ge equency ables, wi h an applica ion o he FDA spon aneous epo ing sys em (wi h discussion) [J]. Am S a , 1999, 53(3):177-190. [19] John Iskande , Vi ali Pool, Weigong Zhou, e al. Da a Mining in he US using he Vaccine Ad e se E en Repo ing Sys em[J]. D ug Sa e y,2006,29(5):375-384. [20] Ba ba a A. Slade, Lau a Leidel, Claudia Vellozzi, e al. Pos licensu e Sa e y Su eillance o Quad i alen Human Papilloma i us Recombinan Vaccine. JAMA 2009; 302(7): 750-757. [21] Ma c suling, I is Pigeo . Signal de ec ion and moni o ing based on longi udinal heal hca e da a, Pha maceu ics 2012, 4, 607-640. [22] Wu Wendi, Liu, Dawei, Li Keli, e al. Analysis on Su eillance Da a o Ad e se E en s Following Immuniza ion in China, 2010. Chinese jou nal o Vaccine and Immuniza ion 2012, 18(5): 385-397. [23] Wu Wendi, Li Keli, Zheng Jingshan, e al. Analysis on Su eillance Da a o Ad e se E en s Following Immuniza ion in China, 2011. Chinese jou nal o Vaccine and Immuniza ion 2013, 19(2): 97-109. [24] Wu Wendi, Liu Dawei, Li Keli, e al. Analysis on Su eillance Da a o Ad e se E en s Following Immuniza ion in China，2012. Chinese jou nal o Vaccine and Immuniza ion 2014, 20(1): 1-12. [25] GACVS/WHO. Global Ad iso y Commi ee on Vaccine Sa e y, 12-13 June 2013. WER, 2013,88(29):301-312. [26] WHO. A ailable om URL: (h p://www.who.in /immuniza ion_s anda ds/ accine_ egula ion/n a_china_ unc ional/en/accessed 15.9.2015) [27] WHO. A ailable om URL: (h p://www.who.in /mediacen e/news/ eleases/2013/japanese_encephali is_20131009/en/ accessed 14.9.2015). [28] Man ed Hauben, Da id Madigan, Cha les M Ge i s. The ole o da a mining in pha maco igilance.Expe Opin. D ug sa 2005, 4(5): 929-948. [29] S ephenson WP, Hauben M. Da a mining o signals in spon aneous epo ing da abases: p oceed wi h cau ion. Pha macoepidemiol D ug Sa 2007;16(4):359–65. [30] Azha Daoud. Feb ile con ulsion: e iew and upda e. Jou nal o Pedia ic Neu ology 2004; 2(1): 9-14. [31] Tonia Jones, S e en J. Jacobsen. Childhood eb ile seizu es: o e iew and implica ions. In e na ional jou nal o medical science 2007, 4(2): 110-114. [32] C Wa ui u, R Apple on. Feb ile seizu es: an upda e. A ch Dis Child 2004;89:751–756. [33] Fe nando Cendes, Raman Sanka . Vaccina ions and eb ile seizu es. Epilepsia 2011, 52(Suppl. 3):23–25.