Effectiveness of 2008--09 Trivalent Influenza Vaccine Against 2009 Pandemic Influenza A (H1N1) --- United States, May--June 2009

Since first reports in April 2009 (1), the 2009 pandemic influenza A (H1N1) virus has spread around the world (2). The pandemic virus is antigenically distinct from seasonal influenza A (H1N1) viruses targeted by seasonal influenza vaccines. Results from recent serologic studies have suggested that seasonal influenza vaccines are unlikely to provide substantial cross-protection against infection with the pandemic H1N1 virus (3). However, how serologic results correlate with the complex immune responses that confer clinical protection remains uncertain. To complement the serologic studies and evaluate the effectiveness of 2008--09 trivalent seasonal influenza vaccine against laboratory-confirmed pandemic influenza A (H1N1) illness, CDC used available data to conduct a case-cohort analysis. The analysis used surveillance reports from eight states of persons aged >18 years with confirmed pandemic H1N1 illness during May--June 2009. Influenza vaccination coverage estimates for these states during the 2008--09 influenza season (September 2008--February 2009) were estimated for the population cohort by using preliminary Behavioral Risk Factor Surveillance Survey (BRFSS) data (4). The overall vaccine effectiveness (VE) against pandemic virus illness after adjustment for age group and presence of chronic medical conditions that increase the risk for complications from influenza was -10% (95% confidence interval [CI] = -43%--15%). Current evidence from this study and other studies does not suggest that seasonal influenza vaccination either decreases of increases the risk for acquiring pandemic H1N1 illness. To prevent seasonal and pandemic influenza, CDC recommends vaccination with seasonal and pandemic influenza vaccines.

The case-cohort method produces a vaccine exposure odds ratio, which for this analysis was an estimate of the relative risk (RR) for 2009 pandemic influenza A (H1N1) illness given seasonal influenza vaccination versus no seasonal vaccination. To obtain the vaccine exposure odds ratio, the odds of vaccination among pandemic H1N1 cases was divided by the odds of vaccination among the population as estimated from BRFSS data. Pandemic H1N1 cases were reported to CDC as part of national outbreak surveillance. The percentage of persons with self-reported seasonal influenza vaccination (receipt of vaccine during September 2008--March 2009) among patients with laboratory-confirmed 2009 pandemic influenza A (H1N1) whose cases were identified in eight states during May--June 2009 was compared with population estimates of vaccination coverage in these states. Only cases of pandemic H1N1 diagnosed in persons aged >18 years in a state providing greater than five reports and with complete patient information on date of birth, illness onset date, presence of a chronic medical condition that increases the risk of influenza complications, and vaccination status were eligible for inclusion in this study. Out of 941 cases in this convenience sample, 356 (38%) had all necessary data available. The 356 case-patients resided in eight states: Arizona (55 patients), Colorado (11), Connecticut (19), Delaware (27), Kentucky (13), Pennsylvania (30), Texas (187), and Virginia (14). For this analysis, laboratory-confirmed 2009 pandemic influenza A (H1N1) infection was defined as a positive test result at state public health laboratories or at CDC by using real-time reverse transcription--polymerase chain (rRT-PCR) protocols, probes, primers, and reagents approved by CDC.

Vaccination coverage for persons aged 18--29 years, 30--39 years, 40--49 years, and >50 years was estimated for the eight selected states by using preliminary 2009 BRFSS data from a telephone survey of 20,689 respondents. Previous BRFSS estimates of vaccine coverage demonstrate that >98% of influenza vaccination occurs before March of the influenza season (CDC, unpublished data, 2009). BRFSS respondents were considered vaccinated if they 1) said "yes" to either having an influenza shot or nasal spray during the past 12 months, and 2) indicated a month and year of vaccination during September 2008--February 2009. Five percent of respondents had unknown influenza vaccination status (i.e., don't know, refused, missing, blank, or incomplete date of vaccination). Because BRFSS does not routinely collect vaccination status on children aged <18 years and uses residential landline telephone numbers, analyses were limited to noninstitutionalized adults aged >18 years (4).

Vaccination coverage estimates were adjusted by four age groups and by the presence of a chronic medical condition that increases the risk for complications from influenza. For all states except Texas, the case surveillance forms recorded whether the patient had any of the following conditions: asthma, chronic heart or circulatory disease, metabolic disease including diabetes, or cancer in the last 12 months. In Texas, the surveillance forms recorded whether the patient had any chronic health condition. The chronic medical conditions for cases were selected to be consistent with those measured by BRFSS, in which survey respondents are asked whether they have ever been told by a doctor, nurse, or other health professional that they have or still have asthma, heart attack, angina, coronary heart disease, stroke, diabetes, or cancer.

Among pandemic H1N1 patients in the analysis, 28% had a chronic medical condition as defined by case surveillance forms, whereas an estimated 22% of the adult population in the BRFSS data from the eight states had at least one of the indicated chronic medical conditions. Within age groups, case and cohort vaccination coverage estimates were adjusted for chronic medical conditions that increase the risk for complications from influenza ("yes" response versus "no") by weighting the stratum-specific estimates by number of cases. Vaccine effectiveness was calculated as 1 -- RR, where RR was the estimated adjusted relative risk for pandemic H1N1 illness as a function of seasonal vaccination coverage. Relative risks were weighted according to the inverse variances of the stratum-specific log RRs. Appropriate statistical software was used to estimate the 2009 BRFSS stratum-specific vaccination coverage for these eight states.

The overall adjusted VE against pandemic virus illness was -10 (CI = -43%--15%). Estimates of VE varied by age group, ranging from -57% to 15% (Table); the CIs for each age group--specific VE estimate were wider than for the overall VE because of reduced sample sizes within age strata.

Reported by: P Gargiullo, PhD, D Shay, MD, J Katz, PhD, A Bramley, MPH, M Nowell, MPH, J Michalove, MPH, L Kamimoto, MD, Influenza Div, JA Singleton, MS, PJ Lu, PhD, MD, Immunization Svc Div, National Center for Immunization and Respiratory Diseases; L Balluz, ScD, Div of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion; A Siston, PhD, EIS Officer, CDC.

Editorial Note:

These results, taken together with other studies, do not support an effect of seasonal 2008--09 trivalent influenza vaccine in either decreasing or increasing the risk for pandemic influenza A (H1N1). The results are consistent with U.S. serologic and immunologic data (3) and with findings from a recently published study from Australia (5). In the immunologic analyses, prevaccination and postvaccination sera from recipients of seasonal influenza vaccines during 2005--2009 were tested by microneutralization methods for levels of cross-reactive antibody to 2009 pandemic influenza A (H1N1) virus. After seasonal vaccination during the 2005--06, 2006--07, and 2008--09 influenza seasons, children aged <10 years lacked detectable neutralizing cross-reactive antibody to the 2009 pandemic influenza A (H1N1) virus. Among adults aged >18 years, vaccination with the 2007--08 or 2008--09 trivalent inactivated vaccine provided little or no increase in cross-reactive antibody levels (3).

In Australia, investigators conducted a case-control study using data from sentinel influenza surveillance practices to assess the effect of seasonal vaccine (5). In-house rRT-PCR assays were used to identify 212 patients with pandemic H1N1 influenza and 365 control patients who tested negative for influenza virus infection. The investigators found no evidence that receipt of seasonal influenza vaccine influenced the risk for being diagnosed with 2009 pandemic influenza A (H1N1) virus infection in any age group (0--4, 5--19, 20--49, 50--64, and >65 years). The overall age-adjusted VE against pandemic virus illness was 3% (CI = -56%--40%).

Findings from other studies examining the effects of 2008--09 influenza vaccine on the risk for pandemic H1N1 virus infection are available. Investigators recently reported results from a hospital-based case-control study conducted in Mexico (6). They reported protection from the 2008--09 trivalent inactivated vaccine against 2009 pandemic influenza A (H1N1) illness. In this study, 60 patients with rRT-PCR--confirmed 2009 pandemic influenza A (H1N1) were frequency matched by age and socioeconomic status to 180 controls examined at the same respiratory disease medical institution (6). The authors reported a vaccine effectiveness of 73% (CI = 34%--89%). However, the authors noted that controls had a higher prevalence of chronic conditions compared with population estimates, thereby likely resulting in a higher vaccination coverage level than the source population. In addition, a series of five studies conducted in four Canadian provinces reportedly found that receipt of seasonal 2008--09 influenza vaccine was associated with a 1.5- to 2-fold greater risk for medically attended 2009 pandemic influenza A (H1N1) illness (7); however, these studies have not yet been published.

Another unpublished study used influenza-like illness (ILI) for its case definition in examining the effect of receipt of 2008--09 seasonal influenza vaccine on the risk for 2009 pandemic influenza A (H1N1). After a large secondary school in New York City experienced an outbreak of ILI, defined as fever (temperature unspecified) with sore throat or cough in April 2009, all students were asked to participate in an online survey assessing ILI and history of influenza vaccination after October 1, 2008. A total of 2,008 (75%) of 2,686 students completed the survey, and 1,607 (60%) students provided both ILI and vaccination status information. Females represented 55% of survey respondents; mean age for both females and males was 15.9 years. Crude, sex-specific, and sex-adjusted relative risks for infection were similar among vaccinated and unvaccinated students, and the overall adjusted RR was 1.05 (CI = 0.91--1.20) (S. Balter, MD, New York City Department of Health and Mental Hygiene, personal communication, 2009).

A case-cohort design was used for the study described in this report. This study design also is known as case-base: vaccination coverage among persons with illness is compared with an estimate of vaccination coverage in the base or source population. This design is similar to the screening method often used to quickly estimate VE in outbreak situations, except that vaccination status is sampled in the population rather than using an assumed true value of the proportion of the population vaccinated (8). A strength of this approach is that it permits rapid estimation of VE after case investigations when existing data on vaccination coverage for the source population is available. A general advantage is that estimating vaccination coverage using a sample from the population rather than from a sample of controls enables dispensing with the rare disease assumption often needed in case-control studies to interpret odds ratios as RRs (9,10). A disadvantage of the stratified case-cohort method used here is that often estimates of population vaccination coverage can be stratified by only a few variables. For example, in this analysis, VE estimates could be stratified only by four age groups, based on the age distribution of the patients and by the presence of a chronic underlying medical condition that increases the risk for complications from influenza. The VE estimates might not have been fully adjusted for age or for the presence of specific conditions, and residual confounding by these factors might be reflected in the results. Also, no adjustment could be done for other possible confounders, such as state of residence, which also might have affected the results.

The findings in this report are subject to at least five other limitations. First, no analysis for children aged <18 years could be performed because limited data were available to determine coverage among children in the 2009 BRFSS. Second, the 2009 pandemic influenza A (H1N1) cases are not necessarily representative of U.S. pandemic influenza cases because they were identified through surveillance in eight states; different levels of case ascertainment also could introduce bias to the extent that vaccination coverage differed among states. Case ascertainment also might be associated with health-seeking behavior and therefore higher levels of vaccination coverage that could have biased these results in the direction of negative VE. Third, the representativeness of the results was affected by using BRFSS coverage estimates, because they are obtained from a landline telephone--only survey of noninstitutionalized persons. Fourth, as with any survey based on self-report of past behavior, a potential for recall bias exists (4). Without record verification of self-reported vaccination status by patients in the study, assessment of recall bias or overreporting bias is difficult, and how such bias might have affected the results is uncertain. Finally, although more than 350 cases were used to estimate the overall VE, the overall CIs are wide, and the CIs for the age group-specific VE estimates are particularly wide, reflecting the smaller sample sizes for these subgroup analyses. Therefore, point estimates, especially the age group--specific estimates, should be interpreted with caution.

This study is part of a growing body of literature examining the effects of seasonal trivalent influenza vaccines on the risk for pandemic H1N1 illness. Taken together, the current evidence does not support a significant effect of 2008--09 trivalent influenza vaccine in either decreasing or increasing the risk for 2009 pandemic influenza A (H1N1) illness. The results from additional studies using more rigorous study designs and methods currently under way in the United States and other countries will further define seasonal influenza VE against pandemic influenza A (H1N1). Studies evaluating the effects of seasonal vaccination on infection with 2009 H1N1 viruses in established animal models for influenza (e.g., ferrets) also are under way at CDC and elsewhere.

CDC and the Advisory Committee on Immunization Practices continue to recommend vaccination with both seasonal and pandemic influenza vaccines to prevent influenza illness during the 2009--10 influenza season in the United States. CDC will continue to monitor the effectiveness of seasonal and pandemic influenza vaccines.

Acknowledgments

This report is based, in part, on contributions by S Lim, MS, D Kapell, MPH, C Zimmerman, MD, T Nguyen, PhD, NYC 2009 Swine Flu Investigation Team, New York City Dept of Health and Mental Hygiene; Arizona Dept of Health Svcs; Colorado Dept of Public Health and Environment; Connecticut Dept of Public Health; Delaware Div of Public Health; Kentucky Dept for Public Health; Pennsylvania Dept of Health; Texas Dept of State Health Svcs; Virginia Dept of Health, and state public health laboratories; J Tate, PhD, Div of Viral Diseases, National Center for Immunization and Respiratory Diseases; W Garvin, C Okoro, MS, Div of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion; and B Silk, PhD, EIS Officer, CDC.

References

1. Novel Swine-Origin Influenza A (H1N1) Virus Investigation Team. Emergence of a novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med. 2009;360:2605--15.
2. Zarocostas J. World Health Organization declares A (H1N1) influenza pandemic. BMJ 2009;338:b2425.
3. Hancock K, Veguilla V, Lu X, et al. Cross-reactive antibody responses to the 2009 pandemic H1N1 influenza virus. N Engl J Med 2009;361 [Epub ahead of print].
4. CDC. Public health surveillance for behavioral risk factor in changing environment. Recommendation from Behavioral Risk Factor Surveillance Team. MMWR 2003;52(No. RR-9).
5. Kelly H, Grant K. Interim analysis of pandemic influenza (H1N1) 2009 in Australia: surveillance trends, age of infection and effectiveness of seasonal vaccination. Euro Surveill 2009;14(31).
6. Garcia-Garcia L, Valdespino-Gomez JL, Lazcano-Ponce E, et al. Partial protection of seasonal trivalent inactivated vaccine against novel pandemic influenza A/H1N1 2009: case-control study in Mexico City. BMJ 2009;339:b3928.
7. Public Health Agency of Canada. Statement on seasonal trivalent inactivated influenza vaccine (TIV) for 2009--2010. Available at http://www.phac-aspc.gc.ca/publicat/ccdr-rmtc/09vol35/acs-dcc-6/index-eng.php. Accessed November 10, 2009.
8. Farrington CP. Estimation of vaccine effectiveness using the screening method. Int J Epidemiol 1993;22:742--6.
9. Knol MJ, Vandenbroucke JP, Scott P, Egger M. What do case-control studies estimate? Survey of methods and assumptions in published case-control research. Am J Epidemiol 2008;168:1073--81.
10. Sato T. Risk ratio estimation in case-cohort studies. Environ Health Perspect 1994;102(Suppl 8):53--6.

Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services. References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of pages found at these sites. URL addresses listed in MMWR were current as of the date of publication.

All MMWR HTML versions of articles are electronic conversions from typeset documents. This conversion might result in character translation or format errors in the HTML version. Users are referred to the electronic PDF version (http://www.cdc.gov/mmwr) and/or the original MMWR paper copy for printable versions of official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to mmwrq@cdc.gov.

Date last reviewed: 11/12/2009