This study examined how public perceptions of influenza and attitudes toward its vaccine changed across three time periods relative to the COVID‑19 pandemic in Yemen, defined in the manuscript as Phase‑I (before COVID‑19), Phase‑II (during COVID‑19), and Phase‑III (after COVID‑19). The primary objective was to assess temporal shifts in public knowledge and vaccine acceptance in a low‑income, fragile nation affected by conflict and constrained health infrastructure.
The investigators distributed a survey questionnaire to members of the public in Yemen. A total of 1,856 respondents completed the survey, allocated to study phases as follows: Phase‑I = 433 participants, Phase‑II = 981 participants, and Phase‑III = 442 participants. The abstract specifies the overall sample sizes by phase but does not report detailed demographic breakdowns, the exact dates or windows defining each phase, nor the full content of the questionnaire in the abstract.
According to the reported results, overall influenza awareness increased in Phase‑II and Phase‑III compared with Phase‑I. The rise in awareness was accompanied by a shift in the sources from which participants obtained information about influenza. The abstract does not provide the detailed breakdown of specific information sources or the relative contribution of each source, but it emphasizes that both awareness levels and information pathways evolved across the pandemic timeline.
Despite the observed increase in general awareness of influenza during and after the COVID‑19 pandemic, the authors report a decline in stated future acceptance of the influenza vaccine. However, the data indicate conditional acceptance increased in later phases: participants in Phases II and III indicated greater willingness to receive the vaccine if certain conditions were met. Those conditions included clear endorsement by the government and by doctors, confirmation that the vaccine was safe and effective, and availability at no cost to recipients through authorities. The abstract does not specify baseline vaccination rates, exact percentages of acceptance or decline, nor whether acceptance differed by demographic groups.
The authors applied logistic regression and machine‑learning models to identify predictors of high influenza awareness and future vaccine acceptance for participants in each phase. The abstract states that these analyses highlighted predictors but does not list the specific predictor variables, effect sizes, model performance metrics, or which predictors were significant in each phase. Therefore, while modelling was performed, the abstract does not provide granular analytic results.
The findings are presented as illustrating the dynamic nature of vaccine behaviour in a fragile, low‑income setting confronting a major pandemic. Key practical implications identified by the authors include:
Authors suggest that these insights can inform the design of targeted vaccination campaigns and the implementation of vaccination policies tailored to the constraints and sociopolitical context of Yemen and similar fragile nations during pandemics and health crises.
The abstract does not report several methodological details that would be relevant for interpretation, including the specific survey items and scoring, demographic and socioeconomic characteristics of respondents, sampling strategy, response rates, timing criteria for phase classification, and the precise predictors identified by the regression and machine‑learning analyses. Effects sizes, confidence intervals, and statistical performance metrics are not provided in the abstract. If required, these details should be sought in the full text.
In summary, the survey of 1,856 participants in Yemen found that awareness of influenza increased during and after the COVID‑19 pandemic, but stated willingness to accept a future influenza vaccine declined overall. Conditional factors—official endorsement, clinician recommendation, proven safety and effectiveness, and free provision—were associated with greater willingness to vaccinate in later phases. Logistic regression and machine‑learning approaches were used to identify predictors for awareness and acceptance, although the abstract does not list those predictors. The authors frame these results as actionable for health authorities planning vaccination campaigns and policies in fragile settings during pandemics.
Note: The abstract summarizes methods and high‑level results but does not provide full analytic details, numeric effect estimates, or the survey instrument. Those specifics were not reported in the abstract and should be consulted in the full article for operational planning or detailed interpretation.