People living within the same household as someone ill with influenza are at increased risk of infection. Here, we use Markov chain Monte Carlo methods to partition the hazard of influenza illness within a cohort into the hazard from the community and the hazard from the household. During the 2013-2014 influenza season, 49 (4.7%) of the 1044 people enrolled in a community surveillance cohort had an acute respiratory illness (ARI) attributable to influenza. During the 2014-2015 influenza season, 50 (4.7%) of the 1063 people in the cohort had an ARI attributable to influenza. The secondary attack rate from a household member was 2.3% for influenza A (H1) during 2013-2014, 5.3% for influenza B during 2013-2014, and 7.6% for influenza A (H3) during 2014-2015. Living in a household with a person ill with influenza increased the risk of an ARI attributable to influenza up to 350%, depending on the season and the influenza virus circulating within the household.

Intra-season timing of influenza infection among persons of different ages could reflect relative contributions to propagation of seasonal epidemics and has not been examined among ambulatory patients. We calculated risk ratios derived from comparing weekly influenza cases pre-peak versus post-peak during the 2010-2011 through 2018-2019 influenza seasons using data from the US Influenza Vaccine Effectiveness network. We sought to determine age specific differences during the ascent versus the descent of a season by influenza virus types and subtypes. We estimated credible intervals around the risk ratios using Bayesian joint posterior sampling of weekly cases. Our population consisted of ambulatory patients with laboratory-confirmed influenza enrolled at five study sites during nine influenza seasons after the 2009 influenza A virus subtype H1N1 (H1N1) pandemic. We observed that young children aged <5 years tended to be more often infected with H1N1 during the pre-peak period while adults aged ≥65 years tended to be more often infected with H1N1 during the post-peak period. However, for influenza A virus subtype H3N2 children aged <5 years were more often infected during the post-peak period. These results may reflect a contribution of different age groups to seasonal spread, which may differ by influenza virus type and subtype.

BACKGROUND: We compared effects of prior vaccination and added or lost protection from current season vaccination among those previously vaccinated. METHODS: Our analysis included data from the US Flu VE Network among participants >/=9 years old with acute respiratory illness from 2012-13 through 2017-18. Vaccine protection was estimated using multivariate logistic regression with an interaction term for effect of prior season vaccination on current season vaccine effectiveness. Models were adjusted for age, calendar time, high-risk status, site, and season for combined estimates. We estimated protection by combinations of current and prior vaccination compared to unvaccinated in both seasons or current vaccination compared to prior vaccinated. RESULTS: 31,819 participants were included. Vaccine protection against any influenza averaged 42% (38 to 47) among those vaccinated only the current season, 37% (33 to 40) among those vaccinated both seasons, and 26% (18 to 32) among those vaccinated only the prior season, compared to participants vaccinated neither season. Current season vaccination reduced the odds of any influenza among patients unvaccinated the prior season by 42% (37 to 46), including 57%, 27% and 55% against A(H1N1), A(H3N2) and influenza B, respectively. Among participants vaccinated the prior season, current season vaccination further reduced the odds of any influenza by 15% (7 to 23), including 29% against A(H1N1) and 26% against B viruses, but not against A(H3N2). CONCLUSION: Our findings support ACIP recommendations for annual influenza vaccination. Benefits of current season vaccination varied among participants with and without prior season vaccination, by virus type/subtype and season.

INTRODUCTION: Several observational studies have shown decreases in measured influenza vaccine effectiveness (mVE) during influenza seasons. One study found decreases of 6%-11% per month during the 2011-12 to 2014-15 seasons. These findings could indicate waning immunity but could also occur if vaccine effectiveness is stable and vaccine provides partial protection in all vaccinees ("leaky") rather than complete protection in a subset of vaccinees. Since it is not known whether influenza vaccine is leaky, we simulated the 2011-12 to 2014-15 influenza seasons to estimate the potential contribution of leaky vaccine effect to the observed decline in mVE. METHODS: We used available data to estimate daily numbers of vaccinations and infections with A/H1N1, A/H3N2 and B viruses. We assumed that vaccine effect was leaky, calculated mVE as 1 minus the Mantel-Haenszel relative risk of vaccine on incident cases and determined the mean mVE change per 30 days since vaccination. Because change in mVE was highly dependent on infection rates, we performed simulations using low (15%) and high (31%) total (including symptomatic and asymptomatic) seasonal infection rates. RESULTS: For the low infection rate, decreases (absolute) in mVE per 30 days after vaccination were 2% for A/H1N1 and 1% for A/H3N2and B viruses. For high infection rate, decreases were 5% for A/H1N1, 4% for A/H3, and 3% for B viruses. CONCLUSIONS: The leaky vaccine bias could account for some, but probably not all of the observed intra-seasonal decreases in mVE. These results underscore the need for strategies to deal with intra-seasonal vaccine effectiveness decline.

The Centers for Disease Control and Prevention recommend annual influenza vaccination of persons >/=6months old. However, in 2016-17, only 43.3% of U.S. adults reported receiving an influenza vaccination. Limited awareness about the cost-effectiveness (CE) or the economic value of influenza vaccination may contribute to low vaccination coverage. In 2017, we conducted a literature review to survey estimates of the CE of influenza vaccination of adults compared to no vaccination. We also summarized CE estimates of other common preventive interventions that are recommended for adults by the U.S. Preventive Services Task Force. Results are presented as costs in US$2015 per quality-adjusted life-year (QALY) saved. Among adults aged 18-64, the CE of influenza vaccination ranged from $8000 to $39,000 per QALY. Assessments for adults aged >/=65 yielded lower CE ratios, ranging from being cost-saving to $15,300 per QALY. Influenza vaccination was cost-saving to $85,000 per QALY for pregnant women in moderate or severe influenza seasons and $260,000 per QALY in low-incidence seasons. For other preventive interventions, CE estimates ranged from cost-saving to $170,000 per QALY saved for breast cancer screening among women aged 50-74, from cost-saving to $16,000 per QALY for colorectal cancer screening, and from $27,000 to $600,000 per QALY for hypertension screening and treatment. Influenza vaccination in adults appears to have a similar CE profile as other commonly utilized preventive services for adults. Efforts to improve adult vaccination should be considered by adult-patient providers, healthcare systems and payers given the health and economic benefits of influenza vaccination.

Background: Vaccination history may confound estimates of influenza vaccine effectiveness (VE) when two conditions are present: (1) Influenza vaccination is associated with vaccination history and (2) vaccination modifies the risk of natural infection in the following seasons, either due to persisting vaccination immunity or due to lower previous risk of natural infection. Methods: Analytic arguments are used to define conditions for confounding of VE estimates by vaccination history. Simulation studies, both with accurate and inaccurate assessment of current and previous vaccination status, are used to explore the potential magnitude of these biases when using different statistical models to address confounding by vaccination history. Results: We found a potential for substantial bias of VE estimates by vaccination history if infection- and/or vaccination-derived immunity persisted from one season to the next and if vaccination uptake in individuals was seasonally correlated. Full adjustment by vaccination history, which is usually not feasible, resulted in unbiased VE estimates. Partial adjustment, i.e. only by prior season's vaccination status, significantly reduced confounding bias. Misclassification of vaccination status, which can also lead to substantial bias, interferes with the adjustment of VE estimates for vaccination history. Conclusions: Confounding by vaccination history may bias VE estimates, but even partial adjustment by only the prior season's vaccination status substantially reduces confounding bias. Misclassification of vaccination status may compromise VE estimates and efforts to adjust for vaccination history.

We read with interest the article by Ray and colleagues and the accompanying editorial by Dr Lipsitch [1, 2]. Ray et al identified a 2-fold increased risk of influenza among individuals vaccinated for >5 months compared to those recently vaccinated. More than a dozen papers have reported findings of declining vaccine effectiveness (VE) with increasing time since vaccination using the test-negative design [3]. We agree with Dr Lipsitch that caution is warranted when inferring that waning immunity is the mechanistic cause of the observed decline in VE.

INTRODUCTION: To evaluate the public health benefit of yearly influenza vaccinations, CDC estimates the number of influenza cases and hospitalizations averted by vaccine. Available input data on cases and vaccinations is aggregated by month and the estimation model is intentionally simple, raising concerns about the accuracy of estimates. METHODS: We created a synthetic dataset with daily counts of influenza cases and vaccinations, calculated "true" averted cases using a reference model applied to the daily data, aggregated the data by month to simulate data that would actually be available, and evaluated the month-level data with seven test methods (including the current method). Methods with averted case estimates closest to the reference model were considered most accurate. To examine their performance under varying conditions, we re-evaluated the test methods when synthetic data parameters (timing of vaccination relative to cases, vaccination coverage, infection rate, and vaccine effectiveness) were varied over wide ranges. Finally, we analyzed real (i.e., collected by surveillance) data from 2010 to 2017 comparing the current method used by CDC with the best-performing test methods. RESULTS: In the synthetic dataset (population 1 million persons, vaccination uptake 55%, seasonal infection risk without vaccination 12%, vaccine effectiveness 48%) the reference model estimated 28,768 averted cases. The current method underestimated averted cases by 9%. The two best test methods estimated averted cases with <1% error. These two methods also worked well when synthetic data parameters were varied over wide ranges (</=6.2% error). With the real data, these two methods estimated numbers of averted cases that are a median 8% higher than the currently-used method. CONCLUSIONS: We identified two methods for estimating numbers of influenza cases averted by vaccine that are more accurate than the currently-used algorithm. These methods will help us to better assess the benefits of influenza vaccination.

BACKGROUND: Estimates of influenza disease burden are broadly useful for public health, helping national and local authorities monitor epidemiologic trends, plan and allocate resources, and promote influenza vaccination. Historically, estimates of the burden of seasonal influenza in the United States, focused mainly on influenza-related mortality and hospitalization, were generated every few years. Since the 2010-2011 influenza season, annual US influenza burden estimates have been generated and expanded to include estimates of influenza-related outpatient medical visits and symptomatic illness in the community. METHODS: We used routinely collected surveillance data, outbreak field investigations, and proportions of people seeking health care from survey results to estimate the number of illnesses, medical visits, hospitalizations, and deaths due to influenza during six influenza seasons (2010-2011 through 2015-2016). RESULTS: We estimate that the number of influenza-related illnesses that have occurred during influenza season has ranged from 9.2 million to 35.6 million, including 140 000 to 710 000 influenza-related hospitalizations. DISCUSSION: These annual efforts have strengthened public health communications products and supported timely assessment of the impact of vaccination through estimates of illness and hospitalizations averted. Additionally, annual estimates of influenza burden have highlighted areas where disease surveillance needs improvement to better support public health decision making for seasonal influenza epidemics as well as future pandemics.

With influenza vaccination rates in the United States recently exceeding 45% of the population, it is important to understand the impact that vaccination is having on influenza transmission. In this study, we used a Bayesian modeling approach, combined with a simple dynamical model of influenza transmission, to estimate this impact. The combined framework synthesized evidence from a range of data sources relating to influenza transmission and vaccination in the United States. We found that, for seasonal epidemics, the number of infections averted ranged from 9.6 million in the 2006-2007 season (95% credible interval (CI): 8.7, 10.9) to 37.2 million (95% CI: 34.1, 39.6) in the 2012-2013 season. Expressed in relative terms, the proportion averted ranged from 20.8% (95% CI: 16.8, 24.3) of potential infections in the 2011-2012 season to 47.5% (95% CI: 43.7, 50.8) in the 2008-2009 season. The percentage averted was only 1.04% (95% CI: 0.15, 3.2) for the 2009 H1N1 pandemic, owing to the late timing of the vaccination program in relation to the pandemic in the Northern hemisphere. In the future, further vaccination coverage, as well as improved influenza vaccines (especially those offering better protection in the elderly), could have an even stronger effect on annual influenza epidemics.

BACKGROUND AND OBJECTIVES: Surveillance for laboratory-confirmed influenza-associated pediatric deaths since 2004 has shown that most deaths occur in unvaccinated children. We assessed whether influenza vaccination reduced the risk of influenza-associated death in children and adolescents. METHODS: We conducted a case-cohort analysis comparing vaccination uptake among laboratory-confirmed influenza-associated pediatric deaths with estimated vaccination coverage among pediatric cohorts in the United States. Case vaccination and high-risk status were determined by case investigation. Influenza vaccination coverage estimates were obtained from national survey data or a national insurance claims database. We estimated odds ratios from logistic regression comparing odds of vaccination among cases with odds of vaccination in comparison cohorts. We used Bayesian methods to compute 95% credible intervals (CIs) for vaccine effectiveness (VE), calculated as (1 - odds ratio) x 100. RESULTS: From July 2010 through June 2014, 358 laboratory-confirmed influenza-associated pediatric deaths were reported among children aged 6 months through 17 years. Vaccination status was determined for 291 deaths; 75 (26%) received vaccine before illness onset. Average vaccination coverage in survey cohorts was 48%. Overall VE against death was 65% (95% CI, 54% to 74%). Among 153 deaths in children with underlying high-risk medical conditions, 47 (31%) were vaccinated. VE among children with high-risk conditions was 51% (95% CI, 31% to 67%), compared with 65% (95% CI, 47% to 78%) among children without high-risk conditions. CONCLUSIONS: Influenza vaccination was associated with reduced risk of laboratory-confirmed influenza-associated pediatric death. Increasing influenza vaccination could prevent influenza-associated deaths among children and adolescents.

In their commentary, Westreich and Hudgens (1) discussed the article by Sullivan et al. (2), in which the authors offered a theoretical framework for examining mechanisms via which bias and confounding may arise in a vaccine effectiveness (VE) study using a test-negative design (TND). We agree that illustrating the TND using directed acyclic graphs provides useful insight, although Sullivan et al. cautioned that they did not address the real-world magnitude or relative importance of the potential biases. Unlike Westreich and Hudgens (1), we found the TND to be valid under a wide range of conditions, a conclusion that is supported by multiple theoretical, simulation-based, and comparative studies (3–5). | To support their opinions, Westreich and Hudgens conducted a simulation study in which individuals with a certain characteristic C = −1 were more likely to be vaccinated (62% vs. 18%). Those with C = −1 could, for example, be people having a pre-existing condition who are specifically targeted for influenza vaccination. In addition, persons in the C = −1 group were much more likely to become infected than were those in the C = 1 group (92% vs. 3% among the unvaccinated). As a result, the risk of influenza infection was greater among vaccinated individuals (23%) than among unvaccinated individuals (7%), and the “crude VE” was strongly negative, illustrating the extreme confounding designed into the example. Using inputs identical to those of Westreich and Hudgens, we found the same likelihood of infection of 11% and the same likelihood of being tested (and therefore observed in the TND study) of 17%. However, in contrast to Westreich and Hudgens, we found a true causal odds ratio of 0.37, not 0.74, using both simulation and calculation (Web Table 1, available at http://aje.oxfordjournals.org/) (6). After adjustment for the characteristic C in a multivariate logistic regression model, we found estimated odds ratios of 0.40 and 0.38 in simulated analyses using the entire cohort and the TND sample, respectively. Thus, even under the extreme conditions of the example used by Westreich and Hudgens, the TND yields an estimated odds ratio similar to the true causal odds ratio, which, unlike the causal relative risk, is independent of C. Finally, given the implausibly high risk of infection among those with C = −1 in this contrived example, the odds ratio is a poor approximation of the relative risk of disease, and 1 minus the odds ratio is not an asymptotically consistent estimator of VE. Fortunately, real-world applications of the TND, which are unlikely to be conducted under such improbable conditions, rarely suffer the same limitation.

Background: Recipients of high-dose vs standard-dose influenza vaccines have fewer influenza illnesses. We evaluated the comparative effectiveness of high-dose vaccine in preventing postinfluenza deaths during 2012-2013 and 2013-2014, when influenza viruses and vaccines were similar. Methods: We identified Medicare beneficiaries aged ≥65 years who received high-dose or standard-dose vaccines in community-located pharmacies offering both vaccines. The primary outcome was death in the 30 days following an inpatient or emergency department encounter listing an influenza International of Classification of Diseases, Ninth Revision, Clinical Modification code. Effectiveness was estimated by using multivariate Poisson regression models; effectiveness was allowed to vary by season. Results: We studied 1039645 recipients of high-dose and 1683264 recipients of standard-dose vaccines during 2012-2013, and 1508176 high-dose and 1877327 standard-dose recipients during 2013-2014. Vaccinees were well-balanced for medical conditions and indicators of frail health. Rates of postinfluenza death were 0.028 and 0.038/10000 person-weeks in high-dose and standard-dose recipients, respectively. Comparative effectiveness was 24.0% (95% confidence interval [CI], .6%-42%); there was evidence of variation by season (P = .12). In 2012-2013, high-dose was 36.4% (95% CI, 9.0%-56%) more effective in reducing mortality; in 2013-2014, it was 2.5% (95% CI, -47% to 35%). Conclusions: High-dose vaccine was significantly more effective in preventing postinfluenza deaths in 2012-2013, when A(H3N2) circulation was common, but not in 2013-2014.

BACKGROUND: The test-negative design (TND) to evaluate influenza vaccine effectiveness is based on patients seeking care for acute respiratory infection, with those who test positive for influenza as cases and the test-negatives serving as controls. This design has not been validated for the inpatient setting where selection bias might be different from an outpatient setting. METHODS: We derived mathematical expressions for vaccine effectiveness (VE) against laboratory-confirmed influenza hospitalizations and used numerical simulations to verify theoretical results exploring expected biases under various scenarios. We explored meaningful interpretations of VE estimates from inpatient TND studies. RESULTS: VE estimates from inpatient TND studies capture the vaccine-mediated protection of the source population against laboratory-confirmed influenza hospitalizations. If vaccination does not modify disease severity, these estimates are equivalent to VE against influenza virus infection. If chronic cardiopulmonary individuals are enrolled because of non-infectious exacerbation, biased VE estimates (too high) will result. If chronic cardiopulmonary disease status is adjusted for accurately, the VE estimates will be unbiased. If chronic cardiopulmonary illness cannot be adequately be characterized, excluding these individuals may provide unbiased VE estimates. CONCLUSIONS: The inpatient TND offers logistic advantages and can provide valid estimates of influenza VE. If highly vaccinated patients with respiratory exacerbation of chronic cardiopulmonary conditions are eligible for study inclusion, biased VE estimates will result unless this group is well characterized and the analysis can adequately adjust for it. Otherwise, such groups of subjects should be excluded from the analysis.

IMPORTANCE: Few studies have evaluated the relationship between influenza vaccination and pneumonia, a serious complication of influenza infection. OBJECTIVE: To assess the association between influenza vaccination status and hospitalization for community-acquired laboratory-confirmed influenza pneumonia. DESIGN, SETTING, AND PARTICIPANTS: The Etiology of Pneumonia in the Community (EPIC) study was a prospective observational multicenter study of hospitalizations for community-acquired pneumonia conducted from January 2010 through June 2012 at 4 US sites. In this case-control study, we used EPIC data from patients 6 months or older with laboratory-confirmed influenza infection and verified vaccination status during the influenza seasons and excluded patients with recent hospitalization, from chronic care residential facilities, and with severe immunosuppression. Logistic regression was used to calculate odds ratios, comparing the odds of vaccination between influenza-positive (case) and influenza-negative (control) patients with pneumonia, controlling for demographics, comorbidities, season, study site, and timing of disease onset. Vaccine effectiveness was estimated as (1 - adjusted odds ratio) x 100%. EXPOSURE: Influenza vaccination, verified through record review. MAIN OUTCOMES AND MEASURES: Influenza pneumonia, confirmed by real-time reverse-transcription polymerase chain reaction performed on nasal/oropharyngeal swabs. RESULTS: Overall, 2767 patients hospitalized for pneumonia were eligible for the study; 162 (5.9%) had laboratory-confirmed influenza. Twenty-eight of 162 cases (17%) with influenza-associated pneumonia and 766 of 2605 controls (29%) with influenza-negative pneumonia had been vaccinated. The adjusted odds ratio of prior influenza vaccination between cases and controls was 0.43 (95% CI, 0.28-0.68; estimated vaccine effectiveness, 56.7%; 95% CI, 31.9%-72.5%). CONCLUSIONS AND RELEVANCE: Among children and adults hospitalized with community-acquired pneumonia, those with laboratory-confirmed influenza-associated pneumonia, compared with those with pneumonia not associated with influenza, had lower odds of having received influenza vaccination.

BACKGROUND: We estimated the burden of outpatient influenza and cases prevented by vaccination during the 2011/2012 and 2012/2013 influenza seasons using data from the United States Influenza Vaccine Effectiveness (US Flu VE) Network. METHODS: We defined source populations of persons who could seek care for acute respiratory illness (ARI) at each of the five US Flu VE Network sites. We identified all members of the source population who were tested for influenza during US Flu VE influenza surveillance. Each influenza-positive subject received a sampling weight based on the proportion of source population members who were tested for influenza, stratified by site, age, and other factors. We used the sampling weights to estimate the cumulative incidence of medically attended influenza in the source populations. We estimated cases averted by vaccination using estimates of cumulative incidence, vaccine coverage, and vaccine effectiveness. RESULTS: Cumulative incidence of medically attended influenza ranged from 0.8% to 2.8% across sites during 2011/2012 and from 2.6% to 6.5% during the 2012/2013 season. Stratified by age, incidence ranged from 1.2% among adults 50 years of age and older in 2011/2012 to 10.9% among children 6 months to 8 years of age in 2012/2013. Cases averted by vaccination ranged from 4 to 41 per 1000 vaccinees, depending on the study site and year. CONCLUSIONS: The incidence of medically attended influenza varies greatly by year and even by geographic region within the same year. The number of cases averted by vaccination varies greatly based on overall incidence and on vaccine coverage.

BACKGROUND: Seasonal influenza causes considerable morbidity and mortality across all age groups, and influenza vaccination was recommended in 2010 for all persons aged 6 months and above. We estimated the averted costs due to influenza vaccination, taking into account the seasonal economic burden of the disease. METHODS: We used recently published values for averted outcomes due to influenza vaccination for influenza seasons 2005-06, 2006-07, 2007-08, and 2008-09, and age cohorts 6 months-4 years, 5-19 years, 20-64 years, and 65 years and above. Costs were calculated according to a payer and societal perspective (in 2009 US$), and took into account medical costs and productivity losses. RESULTS: When taking into account direct medical costs (payer perspective), influenza vaccination was cost saving only for the older age group (65≥) in seasons 2005-06 and 2007-08. Using the same perspective, influenza vaccination resulted in total costs of $US 1.7 billion (95%CI: $US 0.3-4.0 billion) in 2006-07 and $US 1.8 billion (95%CI: $US 0.1-4.1 billion) in 2008-09. When taking into account a societal perspective (and including the averted lost earnings due to premature death) averted deaths in the older age group influenced the results, resulting in cost savings for all ages combined in season 07-08. DISCUSSION: Influenza vaccination was cost saving in the older age group (65≥) when taking into account productivity losses and, in some seasons, when taking into account medical costs only. Averted costs vary significantly per season; however, in seasons where the averted burden of deaths is high in the older age group, averted productivity losses due to premature death tilt overall seasonal results towards savings. Indirect vaccination effects and the possibility of diminished case severity due to influenza vaccination were not considered, thus the averted burden due to influenza vaccine may be even greater than reported.

Mosquito-borne diseases cause significant public health burden and are widely re-emerging or emerging. Understanding, predicting, and mitigating the spread of mosquito-borne disease in diverse populations and geographies are ongoing modelling challenges. We propose a hybrid network-patch model for the spread of mosquito-borne pathogens that accounts for individual movement through mosquito habitats, extending the capabilities of existing agent-based models (ABMs) to include vector-borne diseases. The ABM are coupled with differential equations representing 'clouds' of mosquitoes in patches accounting for mosquito ecology. We adapted an ABM for humans using this method and investigated the importance of heterogeneity in pathogen spread, motivating the utility of models of individual behaviour. We observed that the final epidemic size is greater in patch models with a high risk patch frequently visited than in a homogeneous model. Our hybrid model quantifies the importance of the heterogeneity in the spread of mosquito-borne pathogens, guiding mitigation strategies.

BACKGROUND: Excess mortality due to seasonal influenza is substantial, yet quantitative estimates of the benefit of annual vaccination programs on influenza-associated mortality are lacking. METHODS: We estimated the numbers of deaths averted by vaccination in four age groups (0.5 to 4, 5 to 19, 20 to 64 and ≥65 yrs.) for the nine influenza seasons from 2005/6 through 2013/14. These estimates were obtained using a Monte Carlo approach applied to weekly U.S. age group-specific estimates of influenza-associated excess mortality, monthly vaccination coverage estimates and summary seasonal influenza vaccine effectiveness estimates to obtain estimates of the number of deaths averted by vaccination. The estimates are conservative as they do not include indirect vaccination effects. RESULTS: From August, 2005 through June, 2014, we estimated that 40,127 (95% confidence interval [CI] 25,694 to 59,210) deaths were averted by influenza vaccination. We found that of all studied seasons the most deaths were averted by influenza vaccination during the 2012/13 season (9398; 95% CI 2,386 to 19,897) and the fewest during the 2009/10 pandemic (222; 95% CI 79 to 347). Of all influenza-associated deaths averted, 88.9% (95% CI 83 to 92.5%) were in people ≥65 yrs. old. CONCLUSIONS: The estimated number of deaths averted by the US annual influenza vaccination program is considerable, especially among elderly adults and even when vaccine effectiveness is modest, such as in the 2012/13 season. As indirect effects ("herd immunity") of vaccination are ignored, these estimates represent lower bound estimates and are thus conservative given valid excess mortality estimates.

BACKGROUND: A high-dose trivalent inactivated influenza vaccine was licensed in 2009 by the US Food and Drug Administration (FDA) on the basis of serological criteria. We sought to establish whether high-dose inactivated influenza vaccine was more effective for prevention of influenza-related visits and hospital admissions in US Medicare beneficiaries than was standard-dose inactivated influenza vaccine. METHODS: In this retrospective cohort study, we identified Medicare beneficiaries aged 65 years and older who received high-dose or standard-dose inactivated influenza vaccines from community pharmacies that offered both vaccines during the 2012-13 influenza season. Outcomes were defined with billing codes on Medicare claims. The primary outcome was probable influenza infection, defined by receipt of a rapid influenza test followed by dispensing of the neuraminidase inhibitor oseltamivir. The secondary outcome was a hospital or emergency department visit, listing a Medicare billing code for influenza. We estimated relative vaccine effectiveness by comparing outcome rates in Medicare beneficiaries during periods of high influenza circulation. Univariate and multivariate Poisson regression models were used for analyses. FINDINGS: Between Aug 1, 2012 and Jan 31, 2013, we studied 929 730 recipients of high-dose vaccine and 1 615 545 recipients of standard-dose vaccine. Participants enrolled in each cohort were well balanced with respect to age and presence of underlying medical disorders. The high-dose vaccine (1.30 outcomes per 10 000 person-weeks) was 22% (95% CI 15-29) more effective than the standard-dose vaccine (1.01 outcomes per 10 000 person-weeks) for prevention of probable influenza infections (rapid influenza test followed by oseltamivir treatment) and 22% (95% CI 16-27%) more effective for prevention of influenza hospital admissions (0.86 outcomes per 10 000 person-weeks in the high-dose cohort vs 1.10 outcomes per 10 000 person-weeks in the standard-dose cohort). INTERPRETATION: Our retrospective cohort study in US Medicare beneficiaries shows that, in people 65 years of age and older, high-dose inactivated influenza vaccine was significantly more effective than standard-dose vaccine in prevention of influenza-related medical encounters. Additionally, the large population in our study enabled us to show, for the first time, a significant reduction in influenza-related hospital admissions in high-dose compared to standard-dose vaccine recipients, an outcome not shown in randomised studies. These results provide important new information to be considered by policy makers recommending influenza vaccinations for elderly people. FUNDING: FDA and the office of the Assistant Secretary of Planning and Evaluation.

As influenza vaccination is now widely recommended, randomized clinical trials are no longer ethical in many populations. Therefore, observational studies on patients seeking medical care for acute respiratory illnesses (ARIs) are a popular option for estimating influenza vaccine effectiveness (VE). We developed a probability model for evaluating and comparing bias and precision of estimates of VE against symptomatic influenza from two commonly used case-control study designs: the test-negative design and the traditional case-control design. We show that when vaccination does not affect the probability of developing non-influenza ARI then VE estimates from test-negative design studies are unbiased even if vaccinees and non-vaccinees have different probabilities of seeking medical care against ARI, as long as the ratio of these probabilities is the same for illnesses resulting from influenza and non-influenza infections. Our numerical results suggest that in general, estimates from the test-negative design have smaller bias compared to estimates from the traditional case-control design as long as the probability of non-influenza ARI is similar among vaccinated and unvaccinated individuals. We did not find consistent differences between the standard errors of the estimates from the two study designs.

BACKGROUND: A modification to the case-control study design has become popular to assess vaccine effectiveness (VE) against viral infections. Subjects with symptomatic illness seeking medical care are tested by a highly specific polymerase chain reaction (PCR) assay for the detection of the infection of interest. Cases are subjects testing positive for the virus; those testing negative represent the comparison group. Influenza and rotavirus VE studies using this design are often termed "test-negative case-control" studies, but this design has not been formally described or evaluated. We explicitly state several assumptions of the design and examine the conditions under which VE estimates derived with it are valid and unbiased. METHODS: We derived mathematical expressions for VE estimators obtained using this design and examined their statistical properties. We used simulation methods to test the validity of the estimators and illustrate their performance using an influenza VE study as an example. RESULTS: Because the marginal ratio of cases to non-cases is unknown during enrollment, this design is not a traditional case-control study; we suggest the name "case test-negative" design. Under sets of increasingly general assumptions, we found that the case test-negative design can provide unbiased VE estimates. However, if there were differences in health care-seeking behavior among cases and non-cases by vaccine status, with strong viral interference or when the vaccine modifies the probability of symptomatic illness VE estimates may be biased. CONCLUSIONS: Vaccine effectiveness estimates derived from case test-negative studies are valid and unbiased under a wide range of assumptions. However, if vaccinated cases are less severely ill and seek care less frequently than unvaccinated cases, then an appropriate adjustment for illness severity is required to avoid bias in effectiveness estimates. Viral interference will lead to a non-trivial bias in the vaccine effectiveness estimate from case test-negative studies only when incidence of influenza is extremely high and when duration of transient non-specific immunity is long.