The US Flu VE Network has been described elsewhere [9, 10]. In brief, the public health surveillance network enrolls participants aged ≥8 months with an acute respiratory illness ≤7 days in duration including new or worsening cough at sites in Arizona, Michigan, Missouri, Ohio, Pennsylvania, Texas, and Washington. Duke University, North Carolina, served as the Network’s data coordinating institution. Enrollment began at each site based on local evidence of increasing influenza activity. Trained staff interviewed participants or their parent/guardian to obtain demographic and clinical information and inquired about receipt of current season influenza vaccination. Underlying chronic health conditions were defined as electronic health record (EHR) documentation within the previous year of medical encounters with International Classification of Diseases, Tenth Revision codes for high-risk conditions according to the Advisory Committee on Immunization Practices [1]. Enrollment dates ranged from 1 October 2023 to 30 April 2024 (Supplementary Table 1). The study protocol was reviewed by the US Centers for Disease Control and Prevention (CDC), determined to be public health surveillance, and conducted consistent with applicable federal law and CDC policy (45 CFR 46.102[l][2]).

Northern Hemisphere 2023–2024 quadrivalent influenza vaccines included 2 influenza A antigens and 2 influenza B antigens. Egg-based vaccines were recommended to include A/Victoria/4897/2022-like (H1N1pdm09-like genetic subclade 2a.1)m and cell culture–based (ccIIV4) or recombinant-based (RIV) vaccines were recommended to include A/Wisconsin.67/2022-like (H1N1pdm09-like genetic subclade 2a.1). Each of the 3 vaccine types were recommended to also include A/Darwin/9/2021-like (A/H3N2-like genetic clade 3C.2a1b.2a.2), B/Austria/1359417/2021-like (B/Victoria-like genetic clade V1A.3a.2), and B/Phuket/3073/2013-like (B/Yamagata genetic clade Y3) [11]. Participant influenza vaccination status was determined by combining data from EHRs, state immunization information systems, and plausible self-report. A plausible self-reported influenza vaccination was a dose reported by the participant (or parent/guardian of participant aged 9–17 years) that was received >14 days before illness onset and included a vaccination provider location. All participants were classified as vaccinated if they had received ≥1 dose of any licensed influenza vaccine product after 1 July 2023.

Participants had nasal and oropharyngeal swab specimens obtained (nasal only for children <2 years of age). Specimens were tested for influenza (including type and influenza A subtype) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viruses by means of reverse-transcriptase polymerase chain reaction (RT-PCR) molecular assays at site laboratories. Participants who tested positive for influenza were designated as case patients and participants who tested negative for influenza and SARS-CoV-2 were designated non–case patients (controls). Influenza-positive specimens were genetically characterized using next-generation sequencing locally (1 site, in Arizona) or at CDC (6 sites). Each sample’s hemagglutinin clade was based on its consensus sequence [12].

We excluded the following from the primary analyses: participants with inconclusive influenza test results, participants who tested positive for SARS-CoV-2 [13], and participants who had been vaccinated <14 days before illness onset. Influenza VE was estimated with a test-negative design comparing odds of testing positive for influenza among vaccinated versus unvaccinated participants [14]. VE is expressed as (1 − OR) × 100, where OR is the adjusted odds ratio for influenza among vaccinated persons versus unvaccinated persons from logistic regression models. Models were adjusted a priori for study site, participant age (group for all-age estimates and year of age for age group–specific estimates) at enrollment, presence of ≥1 underlying health condition, and month of illness onset. Additional covariates (ie, sex, race and ethnicity, self-rated general health status, and interval between onset and enrollment) were considered using a threshold for inclusion of ≥5% change in the OR. Estimates include 95% confidence intervals (CIs) where exclusion of 0% indicates statistically significant VE.

We estimated VE by influenza type/subtype and by age group (8 months to 8 years, 9–17 years, 18–49 years, 50–64 years, or ≥65 years). Age group-specific estimates were combined when models did not converge. We also examined VE against A(H1N1)pdm09 among people born in 1958–1979 (aged 45–65 years) because this cohort has been associated with lower effectiveness against A(H1N1)pdm09 in prior seasons [15]. We estimated age group-specific VE against any influenza by time since vaccination using intervals of 14–60, 60–120, and >120 days since vaccination compared with unvaccinated participants.

From 1 October 2023 through 30 April 2024, we enrolled 9061 participants who presented to outpatient and urgent care clinics and emergency departments for acute respiratory illness. We excluded 2432 participants from primary analyses; more than half of excluded participants (1294 [52%]) tested positive for SARS-CoV-2 (Supplementary Figure 1). Of the included 6629 participants, 1780 (27%) tested positive for influenza (Table 1): 806 (45%) for A(H1N1)pdm09, 567 (32%) for influenza B/Victoria, 328 (18%) for A(H3N2), and 104 (6%) for influenza A viruses of undetermined subtype; 20 participants (1%) had influenza A and B virus coinfections, and 5 (<1%) had influenza A(H1N1)pdm09 and A(H3N2) coinfections. Notably, a smaller proportion of B/Victoria cases occurred in participants aged ≥50 years compared with influenza A subtype viruses (Supplementary Figure 2). The number of influenza-positive participants enrolled per week peaked in late January 2024 (Supplementary Figure 3). A total of 2432 participants (37%) were considered vaccinated for influenza for the 2023–2024 season. Of these, 346 (14%) had no documented immunization record of 2023–2024 influenza vaccine but had plausible self-reported receipt.

Influenza-positive case patients were more likely to be male, be younger, identify themselves as black or African American, report better general health, and seek care earlier compared with test-negative controls; case patients were also less likely to have an underlying health condition than controls (Table 1). The proportion of participants vaccinated differed by study site, sex, age, race and ethnicity, general health status, presence of an underlying health condition, and number of days between illness onset and enrollment (Table 1). A total of 1513 vaccinated participants (62%) had a documented vaccine type. Of those with known vaccine type, 1362 (90%) received egg-based vaccines and 151 (10%) received non–egg-based vaccines (123 ccIIV4 and 28 RIV4). Among 654 vaccinated adults aged ≥65 years, 492 (75%) had a known vaccine type. Of those, 467 (95%) received a preferentially recommended vaccine: 281 (57%) received high-dose inactivated influenza vaccine, and 186 (38%) received adjuvanted influenza vaccine; none received RIV4.

A total of 1193 viruses were successfully sequenced: 616 A(H1N1)pdm09, 344 B/Victoria, and 233 A(H3N2). A(H1N1) pdm09 viruses belonged to 2 genetic subclades: 477 (77%) to 6B.1A.5a.2a.1 (2a.1), the same genetic subclade as the 2023–2024 A(H1N1)pdm09 vaccine reference virus, and 139 (23%) to 6B.1A.5a.2a (2a). The 2a.1 genetic subclade predominated among sequenced A(H1N1)pdm09 viruses throughout the enrollment period (Supplementary Figure 4). The distribution of A(H1N1)pdm09 genetic subclades was similar at 6 sites where 2a.1 genetic subclade viruses comprised 76%–86% of sequenced A(H1N1)pdm09 viruses; at the Missouri site, 57% of sequenced A(H1N1)pdm09 viruses belonged to the 2a genetic subclade. All B/Victoria viruses belonged to genetic clade V1A.3a.2, and all A(H3N2) viruses belonged to genetic subclade 2a.3a.1.

From October 2023 through April 2024, the overall adjusted VE against medically attended outpatient influenza A and B viruses was 44% (95% CI, 36%–51%) (Table 2). Estimated VE was 29% (95% CI, 15%–41%) against influenza A(H1N1)pdm09, 74% (65%–81%) against influenza B/Victoria, and 30% (8%–47%) against A(H3N2). VE against A(H1N1)pdm09 varied by age group; VE was lowest, offering no protection against medically attended outpatient illness, among adults aged 50–64 years (−8% [95% CI, −62% to 28%) and was highest among children aged 8 months to 8 years (72% [49%–86%). Among adults born in a specific birth cohort (birth years 1958–1979), we observed no protection in terms of medically attended illness against A(H1N1)pdm09-associated influenza (2% VE [95% CI, −38% to 30%]). Among participants of all ages, VE was higher against influenza A(H1N1)pdm09 genetic subclade 2a viruses (57% [95% CI, 32%–74%]) compared with genetic subclade 2a.1 viruses (26% [7%–41%]) (Figure 1). When stratified by age group, the greatest difference in VE between the 2 genetic subclade was observed among participants aged 18–64 years. VE against A(H1N1)pdm09 genetic subclade 2a.1 viruses in this age group was 13% (95% CI, −16% to 35%) versus 61% (30%–80%) against viruses in the 2a genetic subclade.

The median time between vaccination and illness onset was 94.5 days (interquartile range, 58–134 days]). The interval was shorter among children aged 8 months to 17 years (median [interquartile range], 83 [51–128] days) than among adults aged 18–64 years (95 [56–134] days) and adults aged ≥65 years (99.5 [64.5–142] days). Half of vaccinated adults aged ≥65 years were vaccinated by 4 October 2023, compared with 12 October 2023 for adults aged 18–49 years and 21 October 2023 for children and adolescents aged 8 months to 17 years. Overall and for each age group, VE point estimates were highest among people vaccinated 14–59 days before illness onset compared with those with longer intervals. although CIs largely overlapped (Supplementary Table 2). We observed statistically significant VE against outpatient influenza through ≥120 days after vaccination among both children and adults aged <64 years. Among adults aged ≥65 years, statistically significant VE was not observed ≥60 days after vaccination.

VE calculated using different sources of influenza vaccination data was within 10% points of overall adjusted VE against any influenza combining data sources. VE against any influenza based on self-reported or parent/guardian-reported influenza vaccination was 32% (95% CI, 21%–41%); VE based on EHR-documented data only was 44% (35%–52%) (Supplementary Table 3). Sensitivity analyses varying inclusion criteria for analysis were all within 6% points of the overall adjusted VE against any influenza reported in the primary analysis.