Introduction

Following the 2009 influenza A/H1N1 (pH1N1) pandemic, both seasonal and pH1N1 viruses circulated in the US during the 2010–2011 influenza season; influenza vaccine effectiveness (VE) may vary between live attenuated (LAIV) and trivalent inactivated (TIV) vaccines as well as by virus subtype.

Materials and Methods

Vaccine type and virus subtype-specific VE were determined for US military active component personnel for the period of September 1, 2010 through April 30, 2011. Laboratory-confirmed influenza-related medical encounters were compared to matched individuals with a non-respiratory illness (healthy controls), and unmatched individuals who experienced a non-influenza respiratory illness (test-negative controls). Odds ratios (OR) and VE estimates were calculated overall, by vaccine type and influenza subtype.

Results

A total of 603 influenza cases were identified. Overall VE was relatively low and similar regardless of whether healthy controls (VE = 26%, 95% CI: −1 to 45) or test-negative controls (VE = 29%, 95% CI: −6 to 53) were used as comparison groups. Using test-negative controls, vaccine type-specific VE was found to be higher for TIV (53%, 95% CI: 25 to 71) than for LAIV (VE = −13%, 95% CI: −77 to 27). Influenza subtype-specific analyses revealed moderate protection against A/H3 (VE = 58%, 95% CI: 21 to 78), but not against A/H1 (VE = −38%, 95% CI: −211 to 39) or B (VE = 34%, 95% CI: −122 to 80).

Conclusion

Overall, a low level of protection against clinically-apparent, laboratory-confirmed, influenza was found for the 2010–11 seasonal influenza vaccines. TIV immunization was associated with higher protection than LAIV, however, no protection against A/H1 was noted, despite inclusion of a pandemic influenza strain as a vaccine component for two consecutive years. Vaccine virus mismatch or lower immunogenicity may have contributed to these findings and deserve further examination in controlled studies. Continued assessment of VE in military personnel is essential in order to better inform vaccination policy decisions.

Discussion

The results of this assessment suggest there is a low to moderate degree of protection against A/H3 and B, but not against A/H1 strains that circulated in the US during the 2010–11 season. The low VE against clinically-apparent, laboratory-confirmed influenza illnesses among active component US military service members is somewhat unexpected. However, this is the first study among a primarily US-based population to report VE estimates for the 2010–11 season and may end up being comparable as other data are released on the general US population.

VE estimates for the 2010–11 [12]–[18] and early for 2011–12 season [19], [20] have been reported for the European Union (EU). Although the overall estimates of VE in this study are somewhat lower than those reports from the EU, when the findings are restricted to TIV VE compared to test-negative controls (a more appropriate comparison as the vaccines used in the EU studies are inactivated vaccines), the results are more similar. A study by Kissling et al reported adjusted VE for eight EU states to be 52% overall and 41% for the 15 to 59 year age group [15]. Similar estimates were also reported by Steens et al for the Netherlands (46%) and by Savulescu et al for Spain (50%), both of which used a test-negative control comparison group [13], [14]. Contrary to our findings of no VE for the A/H1 subtype, Kissling et al reported a VE of 27% for A/H1 among 15 to 59 year olds, however, this did not reach statistical significance [15].

There are a number of factors that may have played a role in the low to moderate VE estimates found in this study. There is the potential that the vaccine viruses were a mismatch with the circulating viruses. This has been reported in some previous seasons and has resulted in low VE [24], [25]. Although isolates from the general US population reported by the CDC for the 2010–11 season indicated a close match between the circulating and vaccine viruses [26], this genetic drift could have occurred later in the season and perhaps among strains which circulated among military personnel [27]. This may be especially true for the A/H1 strains where there was an apparent lower immunogenicity and protection provided by vaccines among recruits as described by Myers et al [27]. An additional study by US military collaborators at the US Naval Health Research Center which investigated the genetic characteristics of the A/H1 viruses that circulated in the military recruit population during the 2010–11 season and associated comparisons of the immune responses generated by LAIV and TIV vaccines in this same population is in the publication stage. Noteworthy to mention, however, is the fact that this study has found modest amino acid differences in circulating strains compared to the vaccine strain and could provide much needed answers to these questions (personal communication, Commander Patrick Blair).

Lower than expected VE may also be due to population factors. The military population is highly immunized against influenza, typically at greater than 90% [28]; while the US civilian population of a similar age range (18 to 49 years) has overall vaccination rates of no more than 40% [29]. Previous studies have found decreased VE among highly immunized military populations, especially for the LAIV vaccine, but higher LAIV VE among vaccine-naïve populations, such as military recruits [9]–[11]. For this study, stratification of VE by vaccine type revealed lower and non-significant VE for LAIV recipients compared to TIV. Since almost twice as many of our cases received LAIV compared to TIV, this difference in vaccine type VE may help to explain the overall finding of lower than expected VE in this population. The case-control design of this study may also partially explain the overall lower than expected VE estimates. A simulation model by Ferdinands and Shay, found that case-control studies of VE underestimate true VE by as much as 11.9%, principally due to biases introduced by the lack of diagnostic specificity of tests used (not a factor in our study since we based our cases on RT-PCR and/or culture diagnosis) [30]. All of these explanations warrant additional investigation, perhaps using populations with varying immunization rates and controlled cohort-based studies, to confirm and better understand the mechanisms at play. In addition, VE estimates need to be examined with relation to the degree of severity of influenza-associated illnesses, that is to say, comparison for hospitalized (more severe) versus non-hospitalized outcomes.

One important factor which we could control for was the sensitivity of the influenza-detecting assays given that their sensitivity are known to decrease over time (eg, lower sensitivity of RT-PCR and culture after 48 to 72 hours of illness). In our study, time from symptom onset to specimen collection did not differ between test-positive and test-negative cases (median = 2 days for both groups). Thus, there should have been no difference in influenza detection between test-positive and test-negative cases, given this very narrow sampling window.

There are several strengths and limitations to this study. The use of laboratory-confirmed, clinically-diagnosed influenza cases strengthens this study by providing a more specific case definition. A second strength is the use of both “healthy” and “test-negative” controls for comparison, which provided different methodologies to account for potential biases that can occur in case-control studies of VE [31]. The military population also provides a robust population to study VE as they represent a relatively healthy, young-to-middle aged adult population that is sometimes overlooked in other VE studies. Additionally, medical encounters and vaccines have near complete capture electronically for all active component personnel.

Of note, it is difficult to directly compare the healthy control population to the test-negative control population because the healthy controls were matched to the cases based on demographic characteristics. However, prior history of vaccination does appear to be different between the two control populations. The test-negative controls were more similar to the cases with regards to prior vaccination history (80% with one or more prior influenza vaccinations) than the healthy control population (96%). This probably reflects evidence of better health care seeking behavior and/or opportunities for prior vaccination in healthy controls, thus, comparisons using test-negative controls may represent a more appropriate comparison population in the military population for this and future influenza VE case-control studies.

One important limitation is the fact that the military population is highly immunized, thus, the results of this study may not be generalizable to the general US population. The study was also limited by the number of influenza cases that were laboratory-confirmed. There were probably many more influenza cases that occurred among military personnel, but not all were laboratory-confirmed or sought medical attention. If the cases selected for laboratory confirmation were different from other influenza cases, perhaps due to severity of illness, then the findings may be biased and may not be generalizable to all influenza infections occurring in the military. There may also be unknown biases and confounders that were not accounted for in the adjusted models.

In conclusion, a low level of protection against clinically-apparent, laboratory-confirmed, influenza-associated illness was found for the 2010–11 seasonal influenza vaccines in this military population. TIV immunization was associated with higher protection than LAIV, however, no protection against A/H1 was noted, even though a pandemic virus strain was a vaccine component for the second year in a row. These findings may provide justification towards preferential use of inactivated vaccines as a primary option for “seasoned” (eg, highly-immunized) US military personnel. Continued future annual assessments of influenza vaccine efficacy and/or effectiveness are necessary in the military setting in order to better guide vaccination policies and influenza infection control efforts.

Angelia A. Eick-Cost^1,3*, Katie J. Tastad^2,3, Alicia C. Guerrero², Matthew C. Johns¹, Seung-eun Lee¹, Victor H. MacIntosh², Ronald L. Burke¹, David L. Blazes¹, Kevin L. Russell¹, Jose L. Sanchez^1,3*

1 Armed Forces Health Surveillance Center (AFHSC), Silver Spring, Maryland, United States of America, 2 US Air Force School of Aerospace Medicine (USAFSAM), 711^th Human Performance Wing, Wright Patterson Air Force Base, Ohio, United States of America, 3 Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., (HJF), Bethesda, Maryland, United States of America

Related articles

• No decrease in flu hospitalization from flu shots – Vaccine Journal (therefusers.com)
• 2011 – 12 Seasonal Influenza Vaccines Effectiveness against Confirmed A(H3N2) Influenza Hospitalisation: Pooled Analysis from a European Network of Hospitals. A Pilot Study (plosone.org)