Safety and Effectiveness of a 2009 H1N1 Vaccine in Beijing

Transcription

1 T h e n e w e ngl a nd j o u r na l o f m e dic i n e original article Safety and Effectiveness of a 2009 H1N1 Vaccine in Beijing Jiang Wu, M.D., Fujie Xu, Ph.D., Li Lu, M.Sc., Min Lu, Ph.D., Liang Miao, M.Sc., Ting Gao, M.D., Wenyan Ji, Ph.D., Luodan Suo, M.Sc., Donglei Liu, M.S., Rui Ma, M.Sc., Rui Yu, B.A., Jiazi Zhangzhu, B.A., Weixiang Liu, B.A., Yang Zeng, M.Sc., Xiaomei Li, M.Sc., Xuechun Zhang, B.A., Xinghuo Pang, M.D., and Ying Deng, M.D. A bs tr ac t From the Beijing Centers for Disease Control and Prevention, Beijing (J.W., L.L., M.L., L.M., T.G., W.J., L.S., D.L., R.M., R.Y., J.Z., W.L., Y.Z., X.L., X.Z., X.P., Y.D.); and the Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta (F.X.). Address reprint requests to Dr. Xu at the Centers for Disease Control and Prevention, 1600 Clifton Rd., Mailstop E-02, Atlanta, GA 30333, or at fax1@cdc.gov. Drs. Wu and Xu contributed equally to this article. N Engl J Med 2010;363: Copyright 2010 Massachusetts Medical Society. Background After the first monovalent 2009 pandemic influenza A (H1N1) vaccine became available in September 2009, Chinese officials conducted a mass vaccination program in Beijing. We evaluated the safety and effectiveness of the vaccine. Methods During a 5-day period in September 2009, a total of 95,244 children and adults received the PANFLU.1 vaccine (Sinovac Biotech), a monovalent split-virion vaccine of 15 μg of hemagglutinin antigen without adjuvant. We assessed adverse events after immunization through an enhanced passive-surveillance system and through active surveillance, using diary cards and telephone interviews. Active surveillance for neurologic diseases was implemented in hospitals citywide. To assess vaccine effectiveness, we compared the rates of reported laboratory-confirmed cases of 2009 H1N1 virus infection in students who received the vaccine with the rates in those who did not receive the vaccine, starting 2 weeks after the mass vaccination. Results As of December 31, 2009, adverse events were reported by 193 vaccine recipients. Through hospital-based active surveillance, 362 cases of incident neurologic diseases were identified within 10 weeks after the mass vaccination, including 27 cases of the Guillain Barré syndrome. None of the neurologic conditions occurred among vaccine recipients. From 245 schools, 25,037 students participated in the mass vaccination and 244,091 did not. During the period from October 9 through November 15, 2009, the incidence of confirmed cases of 2009 H1N1 virus infection per 100,000 students was 35.9 (9 of 25,037) among vaccinated students and (687 of 244,091) among unvaccinated students. Thus, the estimated vaccine effectiveness was 87.3% (95% confidence interval, 75.4 to 93.4). Conclusions Among 95,244 children and adults in Beijing, the PANFLU.1 vaccine had a safety profile similar to those of seasonal influenza vaccines and appeared to be effective against confirmed H1N1 virus infection in school-age children. (Funded by the Beijing Municipal Health Bureau.) 2416 n engl j med 363;25 nejm.org december 16, 2010

2 Safety and Effectiveness of an H1N1 Vaccine in Beijing The 2009 pandemic influenza A (H1N1) virus infection was first detected in China in May The first 2009 H1N1 vaccine was licensed in September After the vaccine became available, China s cross-agency expert group, the 2009 H1N1 Influenza Joint Prevention and Control Panel, recommended that the performers and staffers of the 60th National Day Celebration, which was scheduled for October 1, 2009, should be among the first priority groups to receive the vaccine. The celebration was a massive public gathering, and most of the civilian participants were students from primary and middle schools or colleges in Beijing. To address a growing concern about the Guillain Barré syndrome and other adverse events in association with swine influenza vaccines, we greatly enhanced information collection on vaccine safety. The mass vaccination was conducted from September 21 through September 25, This report summarizes the data on vaccine safety and effectiveness from the mass vaccination program. Me thods 2009 H1N1 Vaccine The vaccine used for the mass vaccination program was PANFLU.1 (Sinovac Biotech), a monovalent split-virion vaccine of 15 μg of hemagglutinin antigen without adjuvant. The manufacturing process was the same as those for seasonal influenza vaccines. This H1N1 vaccine was produced from the reassortant strain X-179A (A/ California/07/2009-A/PR/8/34), according to the recommendations of the World Health Organization. 3 The vaccine was purchased by the study sponsor, and the manufacturer was not involved in the study. Vaccine Recipients and Procedures Vaccination was offered to celebration participants who were 4 years of age or older. A preregistration form was used to record vaccine recipients demographic and contact information, including name and telephone number. Personnel at the vaccine sites had to verify all information before administering the vaccine. Because the performers and staffers were organized into groups on different parade floats, the vaccination schedule was drawn up according to participation on each parade float. All vaccine recipients or their guardians provided written informed consent and completed a brief medical history. Vaccination was withheld from participants with a self-reported history of severe reaction to seasonal influenza vaccines, autoimmune diseases, severe allergy, demyelinating diseases, or other serious diseases. Participants who had fever (>37.5 C by selfreport) or acute infectious disease, were pregnant, or had received any other vaccines within the past 30 days were excluded. All vaccine recipients were observed for 30 minutes after being inoculated. A pocket-size reminder card was distributed to all vaccine recipients. The card had instructions on how to report adverse events on one side and telephone numbers of the 18 district- and countylevel offices of the Beijing Centers for Disease Control and Prevention (Beijing CDC) listed on the other side. These telephone numbers connected callers to trained public health professionals who handled calls regarding adverse events after immunization. With the exception of 3 districts or counties, all the centers were open 24 hours a day. This study was approved by the institutional review board of the Beijing CDC. Surveillance for Adverse Events Passive surveillance for adverse events was conducted through the Adverse Event Following Immunization (AEFI) reporting system, which was established in 2005 for the surveillance of vaccine safety after licensure. Anyone, including vaccine recipients and health care providers, can report a suspected AEFI to the Beijing CDC. According to the guidelines for the assessment of adverse events by the State Food and Drug Administration of China, severity was graded on a scale of 1 to 4 as follows: 1, mild symptoms that do not affect daily activities; 2, moderate symptoms that occasionally interfere with daily activities; 3, severe symptoms that affect daily activities or require medication or other treatment; and 4, symptoms that are potentially life-threatening or necessitate a visit to an emergency department or hospitalization. Serious adverse events (grade 3 or 4) are investigated by Beijing s Investigation and Diagnosis Expert Group (BIDEG). This group consists of experienced physicians and experts from many different medical fields, including immunology, allergy, and neurology. All group members have been trained to evaluate AEFIs. Since this was the first mass vaccination against swine influenza, n engl j med 363;25 nejm.org december 16,

3 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 1. Demographic Characteristics of 95,244 Vaccine Recipients. Variable Value no. (%) Occupation College student 45,707 (48.0) Primary- or middle-school student 28,665 (30.1) Teacher, chaperone, or manager 9,036 (9.5) Urban or rural worker 9,441 (9.9) Medical staff 621 (0.7) Retiree or other 1,774 (1.9) Age group 4 11 yr 5,421 (5.7) yr 24,233 (25.4) yr 52,479 (55.1) yr 13,020 (13.7) >60 yr 91 (0.1) BIDEG investigated all reported adverse events that were reported as of December 31, 2009, regardless of severity. Information collected by the Beijing CDC, including demographic data, timing and place of immunization, clinical course of the reported symptoms, use of medical care and clinical diagnoses if any, and medical history, is reviewed by a BIDEG committee of five members. These members typically include two clinicians specializing in the suspected adverse event or disease, an epidemiologist, a public health doctor with experience in administering the vaccine, and a researcher from academia, industry, or a government agency who is knowledgeable about the vaccine product. All vaccine recipients were asked to complete diary cards at 24, 48, and 72 hours and at 7 days after vaccination. The diary cards were distributed at the time of vaccination and were collected by designated persons, according to parade float. Using the diary cards, vaccine recipients could check off local symptoms (pain, redness, swelling, induration, rash, and itching) and systemic symptoms (rash, headache, dizziness, fatigue, sore throat, nausea, vomiting, poor appetite, abdominal pain, diarrhea, muscle pain, joint pain, cough, and fever), along with the severity of symptoms. We placed a telephone call to vaccine recipients who reported any local or systemic symptom to verify the reported symptoms and to encourage those with potentially serious reactions (e.g., fever for ³3 days) to seek medical care. Among vaccine recipients who did not return the diary card, we randomly sampled 20% and interviewed them by telephone. We similarly assessed local and systemic symptoms during the 7 days after vaccination. Surveillance for Neurologic Conditions Active surveillance for neurologic diseases was instituted at all hospitals in the Beijing metropolitan area that have the capacity to diagnose and treat neurologic diseases, totaling 31 specialized and general hospitals. We ascertained the presence of the Guillain Barré syndrome and other selected neurologic diseases, using codes for discharge diagnoses from the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) and manual review of outpatient logs during the 10 weeks after the mass vaccination. Vaccine Effectiveness The student performers who participated in the 60th National Day Celebration came from 245 regular primary and middle schools and many specialized schools. For students from regular schools, the most important criterion for selection was height, so that the participants on each parade float would be of uniform height. To examine the effectiveness of the 2009 H1N1 vaccine, we compared the rates of confirmed cases among students who were vaccinated and those who were not vaccinated. Only students from the 245 regular schools were included in this analysis. Besides the database of the mass-vaccination recipients, we used the school-registration database, which included all students attending the 245 schools, and the database of notifiable infectious diseases in Beijing, which included all reported, confirmed cases of 2009 H1N1 virus infection. In all three databases, the demographic variables to identify individuals include name, sex, age, date of birth, and school. According to China s national guidelines, all patients with confirmed 2009 H1N1 virus infection must have a positive test by the real-time reverse-transcriptase polymerase-chain-reaction (RT-PCR) assay, performed at a laboratory that is operated under the auspices of the Chinese Center for Disease Control and Prevention. The realtime RT-PCR assay was performed in accordance 2418 n engl j med 363;25 nejm.org december 16, 2010

4 Safety and Effectiveness of an H1N1 Vaccine in Beijing Table 2. Adverse Events among 95,244 Vaccine Recipients, According to Age.* Age of Vaccine Recipients No. of Persons Vaccinated Any Adverse Event Vaccination-Related Adverse Event no. of persons % (95% CI) no. of persons % (95% CI) 4 11 yr 5, ( ) ( ) yr 24, ( ) ( ) yr 52, ( ) ( ) yr 13, ( ) ( ) >60 yr All recipients 95, ( ) ( ) * The rates of adverse events and vaccination-related adverse events were reported through the passive surveillance system of the Adverse Event Following Immunization reporting system. Vaccination-related adverse events were adjudicated by Beijing s Investigation and Diagnosis Expert Group. with the protocol from the U.S. CDC. 4 RT-PCR testing was recommended for persons with influenza-like illness who had close contact with a person with confirmed 2009 H1N1 virus infection or who had been exposed to a community in which there was known transmission of 2009 H1N1 virus in the 7 days before the onset of an influenza-like illness and for persons who tested positive for influenza A. For primary and middle school students, public health regulations in Beijing mandated temperature screening when students entered schools in the morning and another temperature screening around noon (implemented from June through November 2009 only). If a student s temperature was 37.5 C (99.5 F) or more, he or she had to seek medical care to rule out 2009 H1N1 virus infection before resuming classes. A total of 18 Beijing CDC laboratories and 37 certified clinical laboratories formed a diagnostic network that processed a large number of pharyngeal or nasopharyngeal swabs daily. Almost all specimens were tested within 3 days after submission. Among all confirmed cases of 2009 H1N1 virus infection that were reported in Beijing, we identified cases from the 245 schools that participated in the mass vaccination. We then determined the vaccination status of each student with confirmed 2009 H1N1 virus infection from the vaccination database. We calculated the incidence per 100,000 students during the period from October 9 (a date that was chosen to give vaccine recipients 14 days after the end of the mass vaccination program for immunity to develop) through November 15, 2009 (which was chosen because starting on November 2, the 2009 H1N1 vaccine became generally available to primary and middle school students in Beijing). Statistical Analysis To estimate vaccine effectiveness, we estimated the relative risk of confirmed 2009 H1N1 virus infection, with effectiveness calculated as (1 relative risk) 100%. We used SPSS software, version 13.0, for all statistical analyses. R esult s Vaccine Recipients A total of 95,244 participants were vaccinated between September 21 and September 25, The age distribution was primarily between 4 and 60 years (Table 1). As of December 31, 2009 (about 14 weeks after vaccination), 193 vaccine recipients reported having adverse events through the AEFI reporting system. Of these adverse events, the most commonly reported systemic symptom was fever (in 90 of 193 recipients). (Details are provided in the Supplementary Appendix, available with the full text of this article at NEJM.org.) The most commonly reported injection-site symptom was pain (in 20 of 193). A total of 137 vaccine recipients with any adverse event (71.0%) reported the onset of symptoms during the first 24 hours after vaccination. BIDEG investigated all 193 reports. Of these, 81 participants were adjudicated to have vaccination-related adverse events, including 3 with serious reactions; all 3 recovered fully within 24 n engl j med 363;25 nejm.org december 16,

5 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Cumulative Incidence (per 100,000) Age in Yr May 13 June 4 June 25 July 16 Aug. 6 Aug. 27 Sept. 17 Oct. 8 Oct. 29 Nov. 19 Dec. 10 Dec. 31 Date of Onset Figure 1. Cumulative Incidence of Confirmed Cases of 2009 H1N1 Virus Infection in Beijing, According to the Date of Symptom Onset and Age. hours without hospitalization. The remaining 78 participants had mild reactions (see the Supplementary Appendix). The overall proportion of vaccine recipients who reported any adverse event was low (0.2%), but there was a significant difference according to age group (P<0.001 for the comparison between participants 4 to 17 years of age and those 18 years of age or older). Children and adolescents between the ages of 4 and 17 years were more likely to report adverse events than participants in other age groups (Table 2). A total of 29,710 diary cards were returned. The compliance rate was 31.2% overall and was highest among children between the ages of 4 and 11 years (54.2%). Among the participants who returned diary cards, 6940 of 29,710 (23.4%) reported having a local or systemic symptom during the 7-day surveillance period. An attempt was made to telephone all participants who reported any symptom, and 4966 of 6940 (71.6%) were reached. Among the 65,534 vaccine recipients who did not return diary cards, attempts were made to telephone a random sample of 13,107 persons, and 11,603 (88.5%) were interviewed. Among those interviewed, 801 of 11,603 (6.9%) reported having a local or systemic symptom during the 7-day surveillance period. Combining diary cards and telephone interviews, we projected that 12.0% of vaccine recipients had either a local or systemic symptom (for details, see the Supplementary Appendix). Surveillance for Neurologic Conditions A total of 362 patients had an incident neurologic condition that met our surveillance criteria, including 27 cases of the Guillain Barré syndrome and 105 cases of optic neuritis (see Section C in the Supplementary Appendix). No match was found between these patients and vaccine recipients in the vaccination database. Vaccine Effectiveness By December 31, 2009, a total of 10,844 confirmed cases of 2009 H1N1 virus infection had been reported in Beijing. The cumulative incidence was highest among adolescents between the ages of 12 and 17 years and lowest among adults 60 years of age or older. The second highest incidence occurred in children between the ages of 4 and 11 years (Fig. 1). The majority of patients had an onset of disease between mid-september and mid-november of To assess vaccine effectiveness in school-age children, we constructed two cohorts: one that included 25,037 primary or middle school students who participated in the mass vaccination program and the other that included 244,091 unvaccinated students from the same schools. The differences with respect to age and sex between the vaccinated cohort and the unvaccinated cohort were not significant, although a significantly greater proportion of vaccinated students lived in an urban district than in a suburban district (Table 3). The difference in the cumulative incidence of 2009 H1N1 virus infection between the vaccinated and unvaccinated cohorts increased over time (Fig. 2). From October 9 through November 15, 2009, a total of 9 confirmed cases of 2009 H1N1 virus infection were reported in the vaccinated cohort and 687 cases in the unvaccinated cohort. The period incidence per 100,000 students was 35.9 (9 of 25,037) in the vaccinated cohort and (687 of 244,091) in the unvaccinated cohort, resulting in a risk ratio of (95% confidence interval [CI], to 0.246). Thus, the vaccine effectiveness was 87.3% (95% CI, 75.4 to 93.4). We did not detect any differences in vaccine effectiveness on the basis of age (P = 0.22 by Fisher s exact test), but the number of cases of 2009 H1N1 virus infection in the vaccinated cohort was small. Indirect Effect of Vaccination Among the 245 schools that participated in the mass vaccination program, the vaccination cov n engl j med 363;25 nejm.org december 16, 2010

6 Safety and Effectiveness of an H1N1 Vaccine in Beijing erage rates varied from less than 1.0% to 54.4%. To assess the indirect effect of vaccination, we compared the period incidence of confirmed cases of 2009 H1N1 virus infection among unvaccinated students on the basis of the school s vaccination coverage. Although the relationship was not linear, the period incidence (from October 9 through November 15, 2009) appeared to be lower in schools with a vaccination coverage of at least 40% than in those with a coverage of less than 40% (P = 0.08 by the chi-square test) (Table 4). Discussion In a previous study, 5 vaccination was shown to be the most efficacious and cost-effective strategy for control of the 2009 H1N1 pandemic. In our study, the PANFLU.1 vaccine had a safety profile similar to those of seasonal influenza vaccines in healthy children and adults. The rate of serious adverse events was low in all age groups, and no potentially vaccine-associated cases of the Guillain Barré syndrome were identified. Our data on vaccine safety were strengthened by active surveillance of adverse events and of serious neurologic conditions from a defined population over a defined time period, by the large size of the population that was studied, and by enhanced passive surveillance. The active surveillance was designed to increase awareness about adverse events and to encourage participants with potentially serious reactions to seek medical care. Hospital-based active surveillance was another strategy we used to detect rare but serious neurologic conditions that might be associated with the influenza vaccine, including the Guillain Barré syndrome and optic neuritis Given the size of the vaccinated population reported here, the study had a statistical power of at least 76% to detect an adverse event occurring at a rate of 1 per 50,000 or more participants, an improvement that adds to the safety data generated from prelicensure vaccine trials. 11 In worldwide safety evaluations of pandemic H1N1 vaccines, it has been shown that these vaccines have an acceptable safety profile. 11,12 The rate of reported adverse events in our study was low, but the rate was higher in children than in adults. This finding is consistent with data from prelicensure trials. 3 The higher rates may be due in part to better compliance in reporting events in children, especially those between the ages of Table 3. Comparison of Vaccinated and Unvaccinated Students between the Ages of 4 and 17 Years at 245 Selected Schools, According to Age Group, Sex, and Residence. Characteristic Cumulative Incidence (per 100,000) Sept. 21 Sept. 27 Oct. 3 Vaccinated (N = 25,037) Oct. 9 number (percent) Oct. 15 Oct. 21 Date of Onset Oct and 11 years, as suggested by the increased rate of diary-card return for this age group. Even though vaccination-related local and systemic reactions may occur, our study provided further assurance that in healthy children and adults between the ages of 4 and 60 years, a nonadjuvanted, inactivated pandemic influenza vaccine had a safety profile similar to those of seasonal influenza vaccines. In our study, the PANFLU.1 vaccine offered substantial protection against laboratory-confirmed 2009 H1N1 virus infection in children and adolescents between the ages of 4 and 17 years. The Unvaccinated (N = 244,091) P Value Age group yr 5,421 (21.7) 53,295 (21.8) yr 19,616 (78.3) 190,796 (78.2) Sex 0.65 Male 12,157 (48.6) 118,887 (48.7) Female 12,880 (51.4) 125,204 (51.3) Residence <0.001 Urban district 24,027 (96.0) 231,731 (94.9) Suburban district 1,010 (4.0) 12,360 (5.1) Unvaccinated students Vaccinated students Nov. 2 Nov. 8 Nov. 14 Figure 2. Cumulative Incidence of 2009 H1N1 Virus Infection in Vaccinated and Unvaccinated Students, According to the Date of Symptom Onset (September 21 November 15, 2009). n engl j med 363;25 nejm.org december 16,

7 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 4. Incidence of Confirmed Cases of 2009 H1N1 Virus Infection among 244,091 Unvaccinated Students at 245 Selected Schools, According to School Vaccination Rate.* School Vaccination Rate No. of Unvaccinated Students No. of Confirmed Cases Period Incidence per 100,000 persons Relative Risk (95% CI) <10% 168, to <20% 49, ( ) 20 to <30% 17, ( ) 30 to <40% 6, ( ) 40 to <50% 1, ( ) 50% ( ) * Only confirmed cases of 2009 H1N1 virus infection that were reported from October 9 through November 15, 2009, were included in this analysis. vaccine effectiveness was consistent with expectations in that for 2009 H1N1 vaccines, the antigenic similarity between vaccine strain and circulating strain was likely to increase vaccine effectiveness. Our findings were also consistent with data from prelicensure trials, 3 in which the seroprotection rate, measured at 21 days, was 81.0% (95% CI, 78.5 to 83.2) in children between the ages of 3 and 11 years and 97.3% (95% CI, 96.1 to 98.1) in adolescents between the ages of 12 and 18 years. 3 Among the 25,037 vaccine recipients at 245 schools in our study (of whom 78.3% were between the ages of 12 and 17 years), we observed a vaccine effectiveness of 87.3% (95% CI, 75.4 to 93.4). Because of the relatively small numbers of participants in each age group, we could not precisely estimate vaccine effectiveness according to specific age. Indirect effects of vaccination are an important consideration when selecting priority groups for vaccination and when setting goals for immunization coverage rates. A stochastic simulation model suggests that the vaccination of 20% of children would reduce the total number of influenza cases in the United States by 46%. 13 In China, because of limited vaccine supply, the target for coverage of the 2009 H1N1 vaccine among school-age children was set at 30%. 2 Our evaluation shows that coverage at such low levels cannot prevent transmission in the population, although our data suggest that the indirect effect of vaccination may have started to appear in schools with vaccine coverage of 30 to 40%. If rates of transmission in households and communities are included, this indirect effect of vaccination is likely to be more important in slowing the transmission and lowering the attack rate within schools and the greater community. The cases of 2009 H1N1 virus infection in Beijing were concentrated among school-age children and adolescents, and the timing of the mass vaccination was before the peak of 2009 H1N1 transmission in the city. Both factors help make our evaluation more informative. A period of reduced incidence was observed during the second week in October 2009 (Fig. 1 and 2). This was probably because a national holiday was observed from October 1 through October 7, and all public schools were closed during the holiday. The temporary decrease in incidence after the national holiday attests to the sensitivity and timeliness of the surveillance system for 2009 H1N1 virus infection in Beijing. Our study has several limitations. First, the diary-card return rate was low, and we had to conduct supplemental surveys among nonrespondents. Second, we assumed that the vaccinated and unvaccinated populations were sufficiently alike in all relevant characteristics other than vaccination to allow the estimation of vaccination effectiveness. The performers of the National Day Celebration were selected on the basis of height, but students who were selected were also required to be physically fit. Therefore, the vaccinated cohort may have been generally healthier than the unvaccinated cohort. Finally, our study population was limited to students attending public primary and middle schools in a large city, and our find n engl j med 363;25 nejm.org december 16, 2010

8 Safety and Effectiveness of an H1N1 Vaccine in Beijing ings may not be generalizable to other populations or in other settings. In conclusion, these data on adverse events should help minimize safety concerns that may be a barrier to the uptake of pandemic influenza vaccines. The vaccine effectiveness of 87.3% against confirmed infection in school-age children is consistent with immunogenicity levels reported from vaccine trials. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the U.S. CDC. Supported by the Beijing Municipal Health Bureau. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. References 1. Cao B, Li XW, Mao Y, et al. Clinical features of the initial cases of 2009 pandemic influenza A (H1N1) virus infection in China. N Engl J Med 2009;361: Stone R. Swine flu outbreak: China first to vaccinate against novel H1N1 virus. Science 2009;325: Liang X-F, Wang H-Q, Wang J-Z, et al. Safety and immunogenicity of 2009 pandemic influenza A H1N1 vaccines in China: a multicentre, double-blind, randomised, placebo-controlled trial. Lancet 2010;375: [Erratum, Lancet 2010;375:1694.] 4. CDC protocol of realtime RTPCR for swine influenza. Geneva: World Health Organization, ( csr/resources/publications/swineflu/cdc realtimertpcrprotocol_ pdf.) 5. Khazeni N, Hutton DW, Garber AM, Owens DK. Effectiveness and cost-effectiveness of expanded antiviral prophylaxis and adjuvanted vaccination strategies for an influenza A (H5N1) pandemic. Ann Intern Med 2009;151: Safety of pandemic vaccines: pandemic (H1N1) briefing note 16. Geneva: World Health Organization, ( briefing_ /en/index.html.) 7. Schonberger LB, Bregman DJ, Sullivan-Bolyaj, et al. Guillain-Barre syndrome following vaccination in the national influenza immunization program, United States, Am J Epidemiol 1979; 110: Safranek TJ, Lawrence DN, Kurland LT, et al. Reassessment of the association between Guillain-Barré syndrome and receipt of swine influenza vaccine in : results of a two-state study. Am J Epidemiol 1991;133: Institute of Medicine. Immunization safety review: influenza vaccines and neurological complications. Washington, DC: National Academy Press, Perry HD, Mallen FJ, Grodin RW, Cossari AJ. Reversible blindness in optic neuritis associated with influenza vaccination. Ann Ophthalmol 1979;11: Pfeifer D, Alfonso C, Wood D. Defining the safety profile of pandemic influenza vaccines. Lancet 2010;375: Girard MP, Katz J, Pervikov Y, Palkonyay L, Kieny MP. Report of the 6th meeting on the evaluation of pandemic influenza vaccines in clinical trials, World Health Organization, Geneva, Switzerland. Vaccine 2010;28: Weycker D, Edelsberg J, Halloran ME, et al. Population-wide benefits of routine vaccination of children against influenza. Vaccine 2005;23: Copyright 2010 Massachusetts Medical Society. receive immediate notification when a journal article is released early To be notified when an article is released early on the Web and to receive the table of contents of the Journal by every Wednesday evening, sign up through our Web site at NEJM.org. n engl j med 363;25 nejm.org december 16,