Emily Toth Martin

Background: Influenza B vaccine effectiveness (VE) has varied by season, lineage, antigenic drift, and vaccine composition. After B/Yamagata detection ceased globally, trivalent vaccines containing B/Victoria were reintroduced in 2024–25. We evaluated influenza B circulation and VE across eight analyzed seasons spanning this transition. Methods: Ambulatory patients aged ≥6 months with acute respiratory illness were enrolled at U.S. Flu VE Network sites during eight influenza seasons from 2016–17 through 2024–25; 2020–21 was excluded because influenza circulation was negligible. Participants were tested by RT-PCR. Lineage-specific VE was estimated using a test-negative design and logistic regression adjusted for age, sex, race/ethnicity, calendar time, and study site. VE was reported for season-lineage strata with ≥50 influenza B cases. Results: Among 65,773 participants, 3,629 influenza B cases were included in lineage-specific VE analyses (1,507 B/Yamagata; 2,122 B/Victoria). Six of 10 season-lineage strata met the reporting threshold. B/Victoria VE was 58% (95% CI, 28%–77%) in 2016–17, 51% (43%–57%) in 2019–20, 78% (71%–84%) in 2023–24, and 39% (14%–56%) in 2024–25. B/Yamagata VE was 55% (46%–63%) in 2016–17 and 49% (40%–56%) in 2017–18. Conclusions: Influenza vaccines protected against medically attended outpatient influenza B illness, although VE varied by season. Continued VE monitoring and genetic surveillance are needed to detect emerging influenza B viruses and inform vaccine strain selection.

Background: Influenza vaccines have seen many changes over recent years in terms of both policy recommendations and product types. Immune responses and vaccine effectiveness vary by vaccine strain composition and product type. This study aimed to explore this variation on a regional level through examining spatial and temporal autocorrelations of vaccine uptake of different product types within Michigan from 2017 to 2021. Methods: Influenza vaccine administration data from the 2017/18 to 2021/22 seasons were obtained from the Michigan Care Improvement Registry. Population size estimates for 2017 and 2019 were obtained from the Michigan Department of Health and Human Services. Age-standardized and age-stratified vaccination rates were computed. Disparate county-level vaccine uptake clusters were constructed using spatial autocorrelation indicators and spatiotemporal clusters were identified using scan statistics. Linear mixed effect models were used to study the impact of social vulnerability and urbanicity level on seasonal influenza vaccination rate. Results: Vaccination incidence rates for adjuvanted, recombinant, and nasal sprays vaccines and for any vaccines increased significantly through the study period. County-level spatiotemporal clusters of type-specific uptake rates were identified. Most notably, we identified an emerging cluster with high uptake of adjuvanted vaccine radiating from Ottawa County, located in suburban Grand Rapids, and the disappearance of high uptake cluster for high-dose vaccine uptake in the area. Vaccination rates by any product type were found to vary seasonally and showed weak negative association with more rural regions. The overall vaccination rates appear to be lower among counties with higher social vulnerability in certain urbanicity levels, but social vulnerability and urbanicity were not significantly associated with overall vaccination rate during the study period. Conclusions: Vaccine type-specific uptake was spatially and temporally diverse in Michigan between 2017 and 2022. Mid and Northern Michigan had consistently relatively lower vaccination rate compared to other regions in Michigan.

Background: Continued serosurvey for SARS-CoV-2 population immunity is critical for vaccine strain selection and identification of susceptible groups such as young children. We aim to evaluate the use of a high-throughput, low-volume assay to describe strain-specific IgG measurement in a longitudinally followed community based cohort. Methods: The longitudinal Household Influenza Vaccine Evaluation study has observed around 300 households annually for acute respiratory illness. SARS-CoV-2 variant-specific IgG (including original, Delta, Beta, and Omicron variants) concentrations were measured using multiplex electrochemiluminescence (ECL), pseudoviral neutralization (PN), and ACE2 inhibition assays in participants who had serum samples drawn between July 1 and December 31, 2021. Spearman’s rank correlation coefficient assessed correlation within and between ECL, ACE2, and PN assays. Receiver operating characteristic (ROC) analysis evaluated the performance of ECL and ACE2 assays. Generalized additive mixed models estimated population immunity by age against SARS-CoV-2 variants. Results: A total of 156 serum samples from 117 participants were selected for this study. Most (65.8%) were from adults and a majority (77.8%) received vaccination during the analysis period. High correlations were observed within ECL IgG assays (0.95–1.00), within ACE2 inhibition assays (0.70–0.99), and between assay platforms, including ECL versus PN (0.81–0.83) and ACE2 versus PN (0.88–0.95). ROC analysis demonstrated strong discrimination of vaccination status by ECL IgG assays (AUC: 0.84–0.88; sensitivity: 0.88–0.91; specificity: 0.87–0.88). ACE2 inhibition assays showed more variable performance (AUC: 0.53–0.96; sensitivity: 0.17–0.93; specificity: 0.91–1.00), with lower performance observed for some BA.2-related antigens. SARS-CoV-2 IgG concentrations generally increased with age and appeared to plateau in early adulthood, although estimates were imprecise in younger age groups because of small sample sizes. Conclusion: We found a high correlation between ECL, ACE2, and PN across variants. Age-specific IgG quantities reflected behavior and infection histories collected by active surveillance.

Infectious disease forecasting in novel outbreaks or low-resource settings is hampered by the need for large disease and covariate datasets, bespoke training, and expert tuning, all of which can hinder rapid generation of forecasts for new settings. To address these challenges, we developed Mantis, a foundation model trained entirely on mechanistic simulations, which enables out-of-the-box forecasting across diseases, regions, and outcomes, even in settings with limited historical data. We evaluated Mantis against 78 forecasting models across 16 diseases with diverse transmission modes, assessing both point forecast accuracy [mean absolute error (MAE)] and probabilistic performance (weighted interval score and coverage). Despite using no real-world data during training, Mantis achieved lower MAE than all models in the CDC’s COVID-19 Forecast Hub when backtested on early pandemic forecasts which it had not previously seen. Across all other diseases tested, Mantis consistently ranked in the top two models across evaluation metrics. Mantis further generalized to diseases with transmission mechanisms not represented in its training data, demonstrating that it can capture fundamental contagion dynamics rather than memorizing disease-specific patterns. These capabilities illustrate that purely simulation-based foundation models such as Mantis can provide a practical foundation for disease forecasting: general-purpose, accurate, and deployable where traditional models struggle.

Background: The US 2024–2025 influenza season was characterized by sustained elevated activity from November 2024 to April 2025, with circulation of both influenza A(H1N1)pdm09 and A(H3N2), the latter of which included some antigenically drifted viruses. Methods: From 1 October 2024 to 30 April 2025, a multistate respiratory virus surveillance network enrolled adults hospitalized with acute respiratory illness in 26 US medical centers. Influenza vaccine effectiveness (VE) against influenza-associated hospitalization and severe in-hospital outcomes was estimated using a test-negative study. The odds of influenza vaccination among influenza-positive case patients and influenza-negative control patients were compared using multivariable logistic regression; VE was calculated as (1−adjusted odds ratio for vaccination) × 100, expressed as a percent. Results: The 2024–2025 seasonal influenza vaccine was effective against influenza-associated hospitalization (VE: 40% [95% confidence interval (CI): 32%–47%]), consistent across age group and influenza type/A subtype. Influenza vaccination also reduced the overall risk of all severe in-hospital outcomes evaluated, including standard oxygen therapy (VE: 41% [95% CI: 31%–50%]), noninvasive advanced respiratory support (VE: 38% [95% CI: 19%–52%]), invasive organ support (VE: 58% [95% CI: 44%–69%]), intensive care unit admission (VE: 58% [95% CI: 47%–67%]), and death (VE: 52% [95% CI: 18%–71%]) with effectiveness varying by influenza A subtype and age. Conclusions: Influenza vaccination reduced the risk of influenza-related hospitalization and severe in-hospital outcomes in adults during the severe 2024–2025 influenza season compared with those not vaccinated.

To characterize anti-nucleocapsid (anti-N) IgG binding antibody responses following first PCR-confirmed SARS-CoV-2 infections in a highly vaccinated population, we evaluated antibody boosting and durability in a longitudinal cohort in Michigan. Among 113 participants with paired pre- and post-infection serum specimens tested using a multiplex electrochemiluminescence assay, SARS-CoV-2 infection induced substantial boosting of anti-N IgG concentrations; approximately 80% of participants becoming anti-N seropositive following infection, and 81% of participants demonstrating at least a 4-fold rise in antibody concentrations following infection. Population-level analyses of 66 participants with post-infection serum specimens collected 52–120 days after infection demonstrated sustained anti-N IgG concentrations over time, with no evidence of substantial decline. These findings demonstrate robust boosting and sustained infection-induced anti-N IgG responses in a highly vaccinated population through four months following SARS-CoV-2 infection. This supports the use of anti-N antibodies for longitudinal serosurveillance, including those who are vaccinated.

Backfill is the process of revising case data, often by retrospectively assigning or reassigning newly reported cases to earlier symptom onset dates. Time- and space-varying delays in backfill may compromise real-time surveillance and forecasting efforts by obscuring true transmission patterns. Using Michigan COVID-19 case data, we developed a statistical mixture model to describe backfill as well as geographic and temporal variations. The model combined an exponential process (case reporting delay) with a gamma-distributed process (case reassignment to onset date). Parameters were estimated by a regularized maximum likelihood method, and the Akaike Information Criterion was used to determine the necessity of the reassignment component for each date. We estimated the exponential reporting speed over time and space and, if appropriate, the weight, transient peak, and time of case reassignment. We found that case reporting improved over the course of the pandemic: reporting speed increased over time (with substantial day-to-day variation), and case reassignments were processed faster. We also identified potential regional disparities: regions with population densities below 50 people/km2 had slower backfill speeds. These findings provide critical insights into the evolution of case reporting and backfill dynamics that can be leveraged for “nowcasting” models to enhance real-time surveillance data, ultimately improving outbreak preparedness and response.

Background: We assessed associations between antibody concentrations within 7 days of symptom onset and testing positive for influenza virus infection among outpatients enrolled in a test-negative study. Methods: From November 2018 to May 2019, study sites in 5 states obtained serum and respiratory specimens from outpatients aged ≥18 years presenting with acute respiratory illness. Respiratory specimens were tested for influenza virus, and viral clades were identified by means of genomic sequencing. We measured influenza antibody titers against vaccine and circulating viruses by means of hemagglutination inhibition (HI), microneutralization (MN) and neuraminidase inhibition (NAI) assays. Reduction in the odds of influenza-associated illness at increasing HI, MN, and NAI antibody titers was estimated using logistic regression, adjusting for influenza vaccination status and time since the beginning of influenza season. Results: Among 175 patients with influenza virus infection, including 112 with influenza A(H1N1)pdm09 and 63 with A(H3N2) (44 clade 3C.3a), and 130 test-negative control patients, higher HI, MN, and NAI antibody titers against circulating influenza viruses were associated with lower odds of confirmed influenza. The odds of A(H1N1)pdm09 infection were 44% and 54% lower for each 2-fold increase in A(H1N1)pdm09 HI or NAI titer, respectively. The odds of A(H3N2) infection were 49% and 28% lower, respectively, for each 2-fold increase in MN or NAI titer against circulating A(H3N2) virus clade. NAI titers were independently associated with lower odds of influenza A(H1N1)pdm09 and A(H3N2) after controlling for HI titer. Conclusions: Higher influenza antibody titers against circulating viruses were associated with lower likelihood of influenza virus infection among adult patients with acute respiratory illness.

Background: Influenza vaccines are designed to induce hemagglutinin-inhibiting (HAI) antibodies rather than neuraminidase-inhibiting (NAI) antibodies; therefore, NAI antibodies’ protective effects need further investigation. Neuraminidase-inhibiting antibodies’ protective effects, independent of HAI antibodies, have been demonstrated for influenza A viruses, but not influenza B viruses (IBVs). Methods: Employing H6NB reassortant virus targets, we evaluated NAI antibodies’ protective effects against IBV in sera from a prospective influenza vaccine effectiveness study among healthcare personnel (HCP) in Israel. Antibody titers against B/Brisbane/60/2008 and B/Phuket/3073/2013 were quantified by HAI and NAI assays. Matched NAI and HAI antibody titers from baseline and postvaccination sera were analyzed for the 2017–2018 influenza season. Results: Laboratory-confirmed IBV infection was recorded in 64/262 HCP and influenza vaccination in 187/262 HCP. Vaccination did not consistently elicit rising NAI titers against vaccine strain B/Brisbane, with a geometric mean fold rise of 1.1 (95% CI: .9–1.3). However, comparing baseline NAI titers, geometric mean titer ratios were 1.50 (95% CI: 1.0–2.1) times higher against B/Brisbane and 2.5 (95% CI: 1.7–3.6) times higher against circulating B/Phuket in vaccinated versus unvaccinated HCP. Each 2-fold rise in NAI antibody titers reduced the hazard of symptomatic infection by 42% (95% CI: 36%–48%) against B/Phuket and 23% (95% CI: 13%–33%) against B/Brisbane, adjusting for HAI titers. Conclusions: Neuraminidase-inhibiting antibodies were a strong correlate of protection against symptomatic IBV infection, with protective effects independent of those from HAI antibodies. These findings suggest that the use and standardization of neuraminidase content in influenza vaccines could provide broader protection against influenza.

Background: In 2023–2024 influenza season, adults in the 1957–1985 birth cohort had the lowest vaccine effectiveness (VE) against A(H1N1)pdm09 viruses than other age groups, including those in the 1986–2005 birth cohort. We investigated A(H1N1)pdm09 antibody levels in sera collected from influenza cases and controls to explain the birth cohort effect of the low VE. Methods: Acute and convalescent sera were collected from adults (1986–2005 and 1957–1985 birth cohorts) during 2023–2024. Hemagglutination inhibiting (HI) and binding antibodies were assessed against representative A(H1N1) viruses and hemagglutinin antigens. Results: A(H1N1)pdm09-infected cases in the 1957–1985 birth cohort had significantly lower HI antibodies in the acute sera against circulating A(H1N1)pdm09 viruses than 1986–2005 birth cohort. Over 60% of the cases in the 1957–1985 birth cohort had HI antibodies ≤10 to circulating A(H1N1)pdm09 viruses. Nonetheless, vaccinated cases in the same 1957–1985 birth cohort mounted robust antibody responses to A(H1N1)pdm09 virus infection, suggesting low vaccine immunogenicity. Antibody landscape analysis revealed that these patients born in 1957–1985 with the lowest VE were likely primed with USSR/77-like A(H1N1) viruses. Conclusions: HI antibody levels in the acute sera were associated with protection against influenza A(H1N1)pdm09 infection. Improving the immunogenicity of the vaccines could help improve VE and overcome the birth cohort effects.