Correspondence About the Paper “Effectiveness of High‐Dose Influenza Vaccine Against Hospitalisations in Older Adults (FLUNITY‐HD): An Individual Level Pooled Analysis” by Niklas Dyrby Johansen et al., https://doi.org/10.1016/S0140‐6736(25)01742‐8

We commend the ambition of the FLUNITY-HD pooled analysis. Based on two harmonized pragmatic trials, GALFLU and DANFLU-2, conducted in Galicia, Spain, and Denmark, respectively, they represent a major effort to assess comparative influenza vaccine effectiveness in older adults using clinically and publicly relevant endpoints. However, several issues regarding interpretation and relevance warrant closer consideration for balanced contextualization. First, event rates were markedly lower than anticipated (0.73% vs. 1.2% in DANFLU-2; 0.34% vs. 0.8% in GALFLU). As a result, GALFLU was underpowered, and its findings remained exploratory, while DANFLU-2 required sample size expansion. These limitations affected individual study power and the statistical validity of the primary endpoint, raising concerns about population representativeness and the robustness of the pooled estimates. Second, both trials were open-label and relied solely on administrative data, introducing potential biases such as unverified coding, misclassification (e.g., pneumonia hospitalizations were three times higher in Denmark than in Galicia), and variability in country-specific coding practices that may not accurately capture influenza-related outcomes. In our view, although these limitations are acknowledged in the manuscript, they have not been consistently integrated into the discussion and conclusions. Third, despite methodological harmonization and no statistically significant variability by season and geography, several factors indicate heterogeneity. These include marked differences in influenza testing rates, gaps in laboratory confirmation of influenza-coded hospitalizations (98.6% in Denmark vs 66.5% in Galicia), and substantial variations in influenza epidemiology. These discrepancies contributed to divergent rVE estimates: DANFLU-2 reported 5.9% (95% CI –2.1 to 13.3), while GALFLU showed 23.7% (95% CI 6.6 to 37.7) with a pinteraction of 0.055. Including pneumonia hospitalizations further dilutes influenza-specific effects, as many cases may not be influenza-related, making the composite endpoint highly heterogeneous and susceptible to coding variability. The two trials were conducted in different geographies (Galicia, Spain, and Denmark) and across three influenza seasons (2022–2023 to 2024–2025), factors likely contributing to differences in influenza epidemiology (i.e., flu subtypes distribution and level of population immunity). These aspects were not adequately addressed in the final analysis. Overall, these issues weaken confidence in influenza-specific endpoints and underscore the challenge of pooling data across regions and seasons—consistent with WHO guidance cautioning against pooling VE estimates when epidemiology and testing practices differ substantially. Fourth, although pooling aimed to increase statistical power, results remain modest and clinically limited. rVE for hospitalization for influenza or pneumonia was 8.8%, statistically significant but with minimal clinical impact: NNV of 1839 (95% CI 1049–9756), corresponding to a risk reduction of only 0.0544%, meaning only five prevented hospitalizations per 10,000 HD-IIV doses. Compared with previous meta-analyses reporting rVE of 13.4% [1] and 23.5% [2], FLUNITY-HD showed contrasting results, underscoring the need for cautious interpretation of its findings. Despite the large sample size, rVE for all-cause mortality was only 1.2%, showing no significant survival benefit. Notably, superiority testing used one-sided p-values, a common approach when the comparator is a placebo, reducing the statistical threshold for significance. In a setting of small absolute differences, this is an important contextual detail. Fifth, the NNV and the current price gap raise critical questions about budget impact and resource allocation. Preventing a single hospitalization would cost tens of thousands of euros, placing the reported public health benefit in a different perspective. The absence of cost-effectiveness or budget-impact analysis, acknowledged as outside the study scope, remains unaddressed in the manuscript. Sixth, the claim that HD-IIV could prevent 1 million all-cause hospitalizations globally is misleading, as it omits the baseline benefit of influenza vaccination itself. Evidence shows vaccination, regardless of type, substantially reduces flu-related events and consequences, including a 25% reduction in all-cause mortality among cardiovascular patients versus controls [3]. This aligns with the recent ESC consensus recommending influenza vaccination, irrespective of vaccine type, as part of cardiovascular prevention, on top of direct influenza control [4]. Moreover, such a general claim overlooks country-specific differences in vaccine availability and immunization budget. This misleading conclusion risks fostering skepticism about the absolute efficacy of SD-IIV vaccines and further fuels vaccine hesitancy, ultimately increasing healthcare resource consumption due to flu among the unvaccinated. We elaborate more on this concept in Figure 1. It illustrates the impact of influenza vaccination among the elderly adults in Denmark and Slovakia, two countries similar in elderly population numbers but opposite vaccination coverage. Although switching to HD-IIV can prevent additional outcomes, the greatest benefit clearly comes from vaccination itself and increasing vaccine coverage, not from the switch. Lastly, given the manufacturer's substantial involvement in study design and manuscript development, cautious interpretation of the marginal absolute benefits and the reliance on one-sided superiority testing is warranted. To conclude, we advocate for a balanced interpretation of FLUNITY-HD findings. Influenza epidemiology varies widely by season and geography, and these factors must inform vaccine effectiveness assessments. Above all, increasing vaccine uptake among older adults, regardless of type, remains the most effective strategy to reduce influenza burden and its consequences. Miloš Jeseňák: conceptualization, validation, writing – review and editing. David Řezáč: conceptualization, validation, writing – review and editing. Laura Colombo: conceptualization, methodology, writing – original draft, visualization, project administration. The editorial support was provided by Pramod Mallikarjuna from Viatris. This study was funded by Viatris. The authors have nothing to report. Miloš Jeseňák has received consulting fees from Abbvie, CSL Behring, Viatris, Novartis, SwiXX, GSK, MSD, and Takeda. He has received payment or honoraria for lectures, presentations, speakers' bureaus, manuscript writing, or educational events from Abbvie, CSL Behring, Viatris, Novartis, SwiXX, Zentiva, MSD, Takeda, AstraZeneca, ALK Slovakia, and Stallergenes-Greer. He has also received support for attending meetings and/or travel from ALK, Novartis, Takeda, and Stallergenes-Greer. David Řezáč has received consulting fees from Hartmann. He has also received payment or honoraria for lectures, presentations, speakers' bureaus, manuscript writing, or educational events from Moderna, Pfizer, and Roche. Laura Colombo is an employee of Viatris. The data that support this letter to editor are available at https://doi.org/10.1016/S0140-6736(25)01742-8.