Effectiveness of seasonal influenza vaccine against dominant A(H1N1)pdm09 in adults ≥ 60 years: a multicenter test-negative design in Zhejiang, China, 2024–2025

Abstract

Background

Zhejiang Province, China, has implemented a free influenza vaccination program for the elderly population for five consecutive years. However, data on real-world vaccine effectiveness against influenza among the local elderly population during the 2024–2025 influenza season are still lacking.

Methods

A multicenter, retrospective test-negative design (TND) was adopted in this study. Sentinel surveillance networks were established across designated sentinel hospitals in six cities of Zhejiang Province, covering the period from October 2024 to July 2025. On a weekly basis, all or a systematic sample of patients aged ≥ 60 years presenting with influenza-like illness (ILI) were enrolled in the study. Epidemiological data were collected, respiratory specimens were obtained, and laboratory assays were conducted to detect influenza virus types and subtypes.

Results

Of the 4694 enrolled elderly patients with ILI, 1587 (33.81%) were influenza-positive, with A(H1N1)pdm09 being the dominant subtype (93.3%). Vaccination coverage was 32.72% (1536/4694). After adjusting for age, sex, chronic comorbidities, calendar week, and city, the overall VE was 53.3% (95% CI: 45.4–60.2). VE estimates by age group were 56.9% (95% CI: 45.8–65.8) for individuals aged 60–69 years, 51.2% (95% CI: 36.2–62.8) for 70–79 years, and 57.7% (95% CI: 37.2–71.8) for those aged ≥ 80 years. When compared with non-consecutive vaccination in the current season, VE for two, three, and four consecutive seasons was 44.6% (95% CI: 31.1–55.5), 35.5% (95% CI: 17.5–49.7), and 38.6% (95% CI: 18.8–53.6), respectively. Protection waned from 58.0% (95% CI: 37.6–71.7) within 59 days post-vaccination to 42.8% (95% CI: 33.4–50.9) between 60 and 180 days. Furthermore, vaccination provided effective protection against moderate-to-severe influenza (68.8%, 95% CI: 50.0-80.6) and influenza-associated hospitalization (53.9%, 95% CI: 37.3–66.0).

Conclusions

In the 2024–2025 season in Zhejiang, influenza vaccination offered moderate protection against laboratory-confirmed influenza among adults aged ≥ 60 years. While VE waned over time, consecutive vaccination maintained moderate effectiveness. The vaccine also provided substantial protection against moderate-to-severe influenza and hospitalization.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12877-026-08049-7.

Introduction

Influenza is an acute respiratory illness that poses a significant global public health threat and imposes a substantial humanistic and economic burden, particularly on the elderly. Evidence indicates that adults aged 65 years and older are at a heightened risk of experiencing more severe symptoms, prolonged illness, and serious complications from influenza [1]. This elevated vulnerability is primarily attributable to a higher prevalence of underlying chronic medical conditions and immunosenescence (age-related decline in innate and adaptive immune function). Consequently, this population faces a markedly increased risk of severe outcomes following infection, including pneumonia, hospitalization, and death [2].

Vaccination is one of the most effective strategies for preventing influenza, especially in the elderly with immunosenescence and multiple comorbidities. However, its effectiveness is challenged by the virus’s continual antigenic drift and shift. These viral changes drive annual epidemics and necessitate yearly vaccine updates. Consequently, the World Health Organization (WHO) issues annual recommendations for the vaccine composition. Zhejiang Province launched a government livelihood project in 2020, initially providing free seasonal influenza vaccinations to adults aged 70 and over. After running for close to four years, the program was expanded in 2024 to cover individuals aged 60 and above [3]. This policy intervention is particularly critical given China’s aging population. In Zhejiang Province, 27.36% of the registered population was aged ≥ 60 years by the end of 2024 [4], underscoring the growing size of this high-risk group. Despite increasing vaccine use, real-world vaccine effectiveness(VE) against medically attended influenza is not constant and must be evaluated each season in the context of specific circulating viruses.

According to the National Influenza Center of China, the 2024–2025 influenza season in China was predominantly characterized by the circulation of influenza A(H1N1)pdm09 viruses. For the period from 1 October 2024 to 6 July 2025, antigenic analysis revealed that 98.6% of the A(H1N1)pdm09 viruses were antigenically similar to the A/Victoria/4897/2022 virus strain included in the season’s vaccine [5]. In Zhejiang Province, substantial influenza activity was observed, where nearly 95,900 cases of influenza were officially reported in December 2024 alone, accounting for 99% of all Category C notifiable infectious diseases during that period [6]. Against this backdrop of substantial epidemic activity and a high degree of antigenic match, evidence-based assessment of the real-world VE becomes crucial.

The test-negative design (TND) has been established as a robust method for evaluating influenza VE in such real-world settings, effectively minimizing selection biases common in observational studies [7]. However, the VE of the 2024–2025 seasonal influenza vaccine has not been sufficiently investigated specifically for the older adult population in Zhejiang Province. Therefore, this study aimed to estimate the influenza VE against medically attended, laboratory-confirmed influenza among adults aged 60 years and older in Zhejiang Province during the 2024–2025 influenza season, by leveraging data from outpatients and emergency department patients within the provincial influenza sentinel surveillance system.

Methods

Study population and design

We conducted a multicenter, retrospective test-negative case-control study. The sentinel hospitals included in this study were selected from the existing Zhejiang Provincial Respiratory Multipathogen Surveillance Network, which has been conducting active surveillance for acute respiratory infections under the coordination of the Zhejiang Provincial CDC since November 2023. This study was initiated in 2024 and conducted in selected sentinel hospitals located in the Northern (Hangzhou, Jiaxing), Eastern (Ningbo), Southern (Wenzhou), and Midwestern (Jinhua, Quzhou) regions of Zhejiang Province, covering a range of healthcare facilities including general hospitals, community health centers, and township health centers. Surveillance departments included internal medicine outpatient and emergency clinics, as well as fever clinics. The study period spanned from October 2024 to July 2025, with enrollment initiated in each city after the local influenza season had begun and at least 14 days after the launch of the seasonal influenza vaccination campaign. Eligible participants were individuals aged ≥ 60 years who met the case definition for influenza-like illness (ILI), which was an acute onset (within ≤ 7 days) with a measured temperature of ≥ 38.0 °C, accompanied by cough or sore throat. Patients or their legal representatives provided written informed consent. Key exclusion criteria were: having received anti-influenza antiviral medications prior to the visit, residing in a nursing home, being a non-local resident (as the provincial free vaccination program is restricted to elderly individuals with local household registration), missing crucial information (e.g., vaccination status or laboratory results), and having received the influenza vaccine less than 14 days before illness onset.

Data collection and definitions

Influenza seasons were defined according to the China CDC surveillance calendar for the winter epidemic period, spanning from October of one year to April of the following year. The influenza vaccination campaign in Zhejiang Province typically begins in September each year, and vaccinations administered from September onward were attributed to the upcoming influenza season. An individual was considered vaccinated for the current season if they had received any eligible influenza vaccine at least 14 days prior to the onset of the current ILI episode. Individuals who received the vaccine after ILI onset were not considered effectively vaccinated for that episode. Consecutive vaccination was defined as receipt of a seasonal influenza vaccine in two or more successive seasons. For example, an individual was considered to have received two consecutive seasons of vaccination if they were vaccinated in both the 2024/10-2025/4 season and the previous 2023/10-2024/4 season. Similarly, three consecutive seasons required vaccination in the 2024/10-2025/4, 2023/10-2024/4, and 2022/10-2023/4 seasons.

In Zhejiang Province, the government livelihood project provided free seasonal influenza vaccination to older adults, primarily using the inactivated trivalent influenza vaccine (IIV3). The 2024–2025 IIV3 formulation contained the strains recommended by the WHO for the Northern Hemisphere: an A/Victoria/4897/2022 (H1N1)pdm09-like virus, an A/Thailand/8/2022 (H3N2)-like virus, and a B/Austria/1,359,417/2021 (B/Victoria lineage)-like virus. The quadrivalent inactivated influenza vaccine (IIV4) was also available as a self-paid alternative.

Influenza vaccination status was ascertained from the Zhejiang Provincial Immunization Program Information Management System. This is a comprehensive, population-based official registry that records vaccination data from all vaccination providers in the province. The system utilizes individuals’ unique national identification numbers for precise data linkage and retrieval, ensuring a complete and accurate vaccination history for each resident. Epidemiological and clinical data were collected by healthcare workers using a standardized questionnaire. The collected information included demographic characteristics, the presence of underlying chronic medical conditions (e.g., asthma, chronic obstructive pulmonary disease, diabetes, cardiovascular diseases, etc.), clinical symptoms, and date of illness onset.

Follow-up and clinical outcome assessment

Disease progression and clinical outcomes were evaluated through standardized telephone interviews conducted 7 ± 2 days following the initial outpatient visit. Data collected included duration of symptoms and hospitalization status.

The classification of clinical severity is based on the relevant standards stipulated in the “Influenza Diagnosis and Treatment Plan” formulated by the National Health Commission [8]. Moderate illness was defined by the requirement for hospitalization or the presence of specified clinical or radiographic findings. Severe or critical illness was defined by the presence of tachypnea, hypoxemia, respiratory failure, shock, or admission to an intensive care unit.

Laboratory testing

Respiratory specimens (oropharyngeal swabs) were collected from enrolled ILI patients. All specimens were tested for influenza virus by real-time reverse transcription-polymerase chain reaction (RT-PCR). Nucleic acid extraction and amplification were performed following standard protocols. Specimens positive for influenza A virus were further subtyped for A(H1N1)pdm09 and A(H3N2), while influenza B virus-positive specimens were further characterized into B/Victoria and B/Yamagata lineages.

Sample size calculation

Prior to study initiation, we performed a sample size calculation using the standard formula for case-control studies: N =2pq(Ua + Ub)2 / (p1 + p2)2. Based on published reports of influenza vaccine effectiveness in the elderly population in China (VE ≈ 0.3) and an expected vaccination coverage (p ≈ 0.3) [9], with a two-sided significance level of α = 0.05 and power of 1–β = 0.80, we estimated that approximately 666 influenza-positive cases and 2220 ILI patients would be required. Accordingly, each participating city was expected to complete no fewer than 400 ILI case investigations, with at least 120 influenza-positive cases collected per city (including at least 30 positive cases aged ≥ 70 years).

Statistical analysis

Descriptive statistics were used to characterize the study population. Continuous variables were summarized using medians and interquartile ranges (IQRs) and compared between groups using the t-test or Mann-Whitney U test. Categorical variables were presented as counts and percentages and compared using the chi-square test or Fisher’s exact test. Vaccine effectiveness was calculated as VE = (1 – adjusted odds ratio) × 100%. The odds ratio (OR) was derived using multivariable logistic regression models that compared the odds of vaccination between influenza-positive cases and influenza-negative controls. The primary model was adjusted for age, sex, and the presence of any underlying chronic medical condition. Subgroup analyses were performed to assess VE by sex, age group, underlying medical condition status, and influenza vaccination history. Vaccination history was also obtained from the official immunization registry. All statistical analyses were performed using R software (version 4.4.2). A two-sided p-value of < 0.05 was considered statistically significant.

Results

Study population characteristics

During the 2024–2025 influenza season, a total of 4694 patients aged ≥ 60 years meeting the ILI case definition were enrolled from 61 sentinel hospitals across six prefectures in Zhejiang Province over the 40-week surveillance period.The sex distribution was balanced, with 2345 males (49.96%) and 2349 females (50.04%). The median age was 70 years, with the largest proportion being 60–69 years old (48.62%, 2282/4694), followed by 70–79 years (33.98%, 1595/4694), 80–89 years (15.02%, 705/4694), and ≥ 90 years (2.39%, 112/4694). A total of 3,009 participants (64.10%) reported having at least one underlying chronic medical condition.

Most enrolled ILI cases had their onset between December 2024 and March 2025, with the peak number of cases observed in January 2025. Fever (92.0%) and cough (73.6%) were the most common symptoms, followed by sore throat (38.3%) and sputum production (31.9%). Other symptoms including fatigue, rhinorrhea, and headache were reported in less than 20% of cases. Severe symptoms such as dyspnea and chest pain were uncommon (≤ 2.3%). The distribution of clinical symptoms among enrolled ILI cases is shown in Fig. 1.

Fig. 1.

Distribution of clinical symptoms

In terms of clinical severity, a total of 367 (7.8% of all enrolled) patients were hospitalized, of 74.4% (273/367) tested positive for influenza and 26.4% (97/367) had received the current seasonal influenza vaccine. A total of 1289 patients were not hospitalized, and hospitalization status was not recorded for 3038 participants. Regarding clinical severity classification, due to the small number of severe cases, moderate and severe cases were combined for analysis. A total of 179 (3.6% of all enrolled) patients were classified as moderate-to-severe influenza cases (including 174 moderate and 5 severe cases), of 73.7%(132/179) tested positive for influenza and 20.1%(36/179)had received the current seasonal influenza vaccine.

Influenza vaccination status

Among the enrolled participants, 1536 (32.72%) received the seasonal influenza vaccine during the 2024–2025 season. The mean time from vaccination to illness onset was 119 ± 49 days. Vaccinated individuals were significantly younger (median age 68 years) than unvaccinated individuals (median age 73 years). Statistically significant differences were observed between the vaccinated and unvaccinated groups in terms of age, chronic comorbidities, and vaccination history (all p < 0.001). The vast majority of vaccinated participants received the trivalent influenza vaccine (99.3%, 1526/1536), while only 10 received the quadrivalent vaccine. Among the 1021 participants vaccinated in the 2023–2024 season, 68.36% (698/1021) were also vaccinated in the current season, indicating a high rate of vaccination continuation, as shown in Table 1.

Table 1.

Characteristics, vaccination status, and influenza infection among enrolled ILI cases

	Total (n = 4694)	Not Vaccinated (n = 3158)	Vaccinated (n = 1536)	P	Influenza Positive (n = 1587)	Influenza Negative (n = 3107)	P
Age, IQR	70 (64, 77)	68 (63, 75)	73 (67, 79)	< 0.001	67 (62, 74)	71 (65, 78)	< 0.001
Gender, n(%)				0.023			0.001
Male	2345 (49.96)	1541 (65.71)	804 (34.29)		740 (31.56)	1605 (68.44)
Female	2349 (50.04)	1617 (68.84)	732 (31.16)		847 (36.06)	1502 (63.94)
Age Group, n(%)				< 0.001			< 0.001
individuals aged 60–69 years	2282 (48.62)	1714 (75.11)	568 (24.89)		947 (41.50)	1335 (58.50)
70–79 years	1595 (33.98)	974 (61.07)	621 (38.93)		447 (28.03)	1148 (71.97)
80–89 years	705 (15.02)	398 (56.45)	307 (43.55)		162 (22.98)	543 (77.02)
above 90 years	112 (2.39)	72 (64.29)	40 (35.71)		31 (27.68)	81 (72.32)
Chronic comorbidities, n(%)				< 0.001			< 0.001
No	1685 (35.90)	1193 (70.80)	492 (29.20)		641 (38.04)	1044 (61.96)
Yes	3009 (64.10)	1965 (65.30)	1044 (34.70)		946 (31.44)	2063 (68.56)
Vaccination in the previous influenza season(2023–2024), n(%)				< 0.001			< 0.001
No	3673 (78.25)	2835 (77.18)	838 (22.82)		1369 (37.27)	2304 (62.73)
Yes	1021 (21.75)	323 (31.64)	698 (68.36)		218 (21.35)	803 (78.65)
Vaccination in the current influenza season(2024–2025), n(%)							< 0.001
No	3158 (67.28)	-	-		1239 (39.23)	1919 (60.77)
Yes	1536 (32.72)	-	-		348 (22.66)	1188 (77.34)	< 0.001

Influenza positivity and associated factors

We confirmed a total of 1587 influenza-positive cases, corresponding to an overall positivity rate of 33.81%, and found that influenza A(H1N1)pdm09 emerged as the dominant circulating subtype, representing 93.3% (1480/1587) of confirmed influenza-positive cases. The small proportion of remaining positive samples included A(H3N2) (0.1%, 2/1587), unsubtyped influenza A (6.5%, 103/1587), and influenza B/Victoria lineage (0.1%, 2/1587). In terms of temporal distribution, influenza activity exhibited a distinct seasonal pattern. The number of ILI cases and laboratory-confirmed influenza cases increased substantially from late December 2024, peaked in January 2025, and gradually declined thereafter ( Supplementary Fig. 1), with detailed weekly data presented in Supplementary Table 1.

Influenza-positive cases were significantly younger (median age 67 years) than test-negative controls (median age 71 years). Significant differences were observed between positive and negative groups regarding age, sex, chronic comorbidities, and vaccination history (all p < 0.001). The positivity rate was significantly lower among those vaccinated in the current season (22.66%, 348/1536) compared to the unvaccinated (39.23%, 1239/3158; p < 0.001), as shown in Table 1. Multivariable analysis confirmed that vaccination in both the current and previous seasons were independent protective factors against influenza infection. Additionally, female sex, younger age, and absence of chronic comorbidities were associated with a higher risk of influenza infection, as shown in Table 2.

Table 2.

Multivariable analysis of factors associated with influenza infection

Factors	OR (95%CI)	P
Gender (Reference: Male)
Female	1.16 (1.03 ~ 1.32)	0.018
Age Group (Reference: individuals aged 60–69 years)
70–79 years	0.65 (0.56 ~ 0.75)	< 0.001
80–89 years	0.52 (0.43 ~ 0.64)	< 0.001
above 90 years	0.63 (0.41 ~ 0.96)	0.033
Chronic comorbidities (Reference: No)
Yes	0.87 (0.76 ~ 0.99)	0.034
Vaccinationin the previous influenza season(2023–2024) (Reference: No)
Yes	0.55 (0.47 ~ 0.64)	< 0.001
Vaccination in the current influenza season(2024–2025) (Reference: No)
Yes	0.73 (0.61 ~ 0.88)	< 0.001

Influenza vaccine effectiveness

The unadjusted VE was 54.6% (95% CI: 47.9–60.6). After adjusting for age, sex, chronic comorbidities, calendar week, and city using a logistic regression model, the overall influenza VE against laboratory-confirmed influenza among adults aged ≥ 60 years was 53.3% (95% CI: 45.4–60.2), as shown in Fig. 2. The VE against the predominant A(H1N1)pdm09 subtype was comparable to the overall VE, at 53.0% (95% CI: 44.8–60.0).

Fig. 2.

Adjusted influenza VE and 95% CI overall and by subgroup. VE estimates were adjusted for potential confounders including age, sex, chronic comorbidities, calendar week, and city. Estimates for influenza A/H3N2 and B/Victoria were not calculated due to insufficient sample size

In subgroup analyses, VE was 56.1% (95% CI: 45.3–64.9) in females and 51.0% (95% CI: 38.6–61.0) in males. VE estimates by age group were 56.9% (95% CI: 45.8–65.8) for individuals aged 60–69 years, 51.2% (95% CI: 36.2–62.8) for 70–79 years, and 57.7% (95% CI: 37.2–71.8) for ≥ 80 years. VE was 52.1% (95% CI: 41.8–60.6) among those with underlying conditions and 56.1% (95% CI: 42.5–66.6) among those without.

Due to the limited number of cases in these severity subgroups, we used all influenza-negative patients as the common control group, vaccination provided significant protection against moderate-to-severe influenza, with an adjusted VE of 68.8% (95% CI: 50.0-80.6), and against influenza-associated hospitalization, with an adjusted VE of 53.9% (95% CI: 37.3–66.0), see Fig. 3.

Fig. 3.

Adjusted influenza VE against clinical severity outcomes among older adults. VE against moderate-to-severe influenza and hospitalization, with numbers of influenza-positive cases and vaccinated cases for each outcome. All estimates were adjusted for potential confounders

When compared with non-consecutive vaccination in the current season (defined as individuals vaccinated in 2024–2025 but with at least one missing vaccination in preceding seasons), VE was 44.6% (95% CI: 31.1–55.5) for those with two consecutive seasons of vaccination, 35.5% (95% CI: 17.5–49.7) for three consecutive seasons, and 38.6% (95% CI: 18.8–53.6) for four consecutive seasons (Fig. 4a). We further examined whether the effectiveness of consecutive vaccination varied across key subgroups. Interaction tests between consecutive vaccination status (2-year, 3-year, and 4-year) and age group, sex, and comorbidity status were all statistically non-significant (all p for interaction > 0.05). VE was highest shortly after vaccination and declined over time. Compared with unvaccinated individuals, the VE was 58.0% (95% CI: 37.6–71.7) within 59 days post-vaccination, decreasing to 42.8% (95% CI: 33.4–50.9) between 60 and 180 days, see Fig. 4(b). VE beyond 180 days could not be calculated due to limited data (only 2 positive cases).

Fig. 4.

Adjusted influenza VE among aged ≥ 60 years by consecutive vaccination status (a) and by time interval from vaccination to ILI onset (b). (a) VE estimates for individuals with two, three, and four consecutive years of vaccination prior to the season, with corresponding numbers of influenza-positive cases and vaccinated cases. (b) VE estimates according to time elapsed between vaccination and ILI onset, with case numbers for each interval. All estimates were adjusted for potential confounders

Discussion

During the 2024–2025 influenza season in Zhejiang Province, A(H1N1)pdm09 emerged as the predominant circulating subtype, sharing significant antigenic similarity with the A/Victoria/4897/2022 strain included in the seasonal influenza vaccine [5]. In the preceding seasons, influenza activity in China was characterized by different predominant subtypes. The 2022–2023 season was dominated by co-circulation of A(H1N1)pdm09 and A(H3N2), while the 2023–2024 season featured co-circulation of A(H3N2) and B/Victoria lineage viruses. Our study assessed the real-world protective effect of seasonal influenza vaccines among adults aged ≥ 60 years, revealing that vaccination afforded moderate protection against laboratory-confirmed influenza in this population.

A comparative analysis with studies conducted in other regions during the same season revealed both consistencies and variations in VE estimates. A test-negative design study in Beijing, which also documented A(H1N1)pdm09 as the predominant subtype, reported an overall VE of 48.3% in the general population, while the VE for adults aged ≥ 60 years was 31.5%. The relatively lower VE in Beijing’s elderly population may be attributed to differences in regional vaccination coverage, population health status, or confounding factors adjusted in the analysis [10]. Internationally, interim VE data for the same season showed diverse results: Germany reported a VE of 35% among adults ≥ 60 years based on primary care and hospital sentinel surveillance [11], Canada documented a VE of 57% for adults ≥ 65 years [12], and Finland’s cohort study estimated a 50% reduction in influenza risk among those ≥ 65 years [13]. Notably, the United States, where A(H3N2) was predominant, still achieved a VE of 51% (95% CI: 47–54) among adults ≥ 65 years against outpatient medically attended influenza [14]. These discrepancies highlight that VE is influenced by multiple factors, including circulating virus subtypes, vaccine composition, population characteristics, and study design. Overall, the VE estimate from the present study falls within the international range for elderly populations during the same season, confirming the reliability of the finding that the vaccine provided moderate protection against the predominant A(H1N1)pdm09 virus.

In the present study, the adjusted VE for individuals aged ≥ 80 years was slightly higher than that observed in the 60–69 and 70–79 years age groups. Similarly, a recent Italian test-negative design study reported a relative VE of 54% (95% CI: 10%-76%) for high-dose influenza vaccine against laboratory-confirmed influenza among adults aged ≥ 80 years [15], and a population-based cohort study in Spain found that high-dose vaccination was associated with a 43.1% (95% CI: 24.6%-61.7%) relative reduction in laboratory-confirmed influenza among individuals aged ≥ 80 years [16]. Although immunosenescence theory suggests that age-related declines in immune function may impair vaccine responsiveness in older adults [17, 18], these findings indicate that the oldest old do not necessarily exhibit lower VE.

Nevertheless, in our study, the median age of vaccinated participants was significantly lower than that of unvaccinated participants, suggesting the potential presence of healthy vaccinee bias. Observational studies of influenza VE in older adults have long been challenged by frailty selection bias, whereby individuals with higher levels of frailty, due to mobility limitations, cognitive impairment, or functional decline, are less likely to be vaccinated, yet face substantially higher baseline risks of influenza infection and severe outcomes [19]. Unmeasured health-related factors such as frailty and functional dependence may introduce residual confounding and potentially overestimate VE [20]. Future research should pay greater attention to the oldest-old population and explore the feasibility of stratified vaccination strategies. Given that high-dose influenza vaccines have demonstrated superior protection in this age group, recommending higher-dose formulations specifically for individuals aged ≥ 80 years may help optimize protection for this most vulnerable population.

However, with respect to risk factors for influenza positivity among older adults, the multivariable analysis in our study revealed that younger age and the absence of chronic comorbidities were associated with higher infection rates. This pattern likely reflects differences in exposure opportunity. Healthier and relatively younger older adults tend to have more active social lives and more frequent community contact, thereby increasing their opportunities for influenza virus exposure. This finding is consistent with a previous study in Japan, which reported that higher social participation among older adults was associated with a significantly increased risk of self-reported influenza infection [21].

Concurrently, a waning of protection was observed over time post-vaccination. This pattern is consistent with a U.S. study during the 2023–2024 season, reported that VE among adults ≥ 65 years was 63% at 14–59 days post-vaccination, but became non-significant beyond 60 days with a VE of only 15% [22]. These observations align with findings from a large-scale cohort study in Japan, which reported that influenza VE in the elderly wanes progressively, with detectable protection generally maintained for 4 to 5 months post-vaccination [23]. The decline in effectiveness over time may be attributable to factors such as waning immunity and antigenic drift of circulating viruses. These findings collectively reinforce the importance of early seasonal vaccination to ensure optimal protection in the elderly. To this end, it is recommended that elderly individuals, particularly those aged ≥ 80 years with multiple comorbidities, receive influenza vaccination 1–2 months prior to the peak influenza season to mitigate the risk of waning immunity during the epidemic period.

The impact of repeated influenza vaccination on VE remains a subject of discussion [24]. In our study, when consecutive vaccinees were compared with non-consecutive vaccinees in the current season, the point estimates suggested a numerically lower VE with longer-term consecutive vaccination. However, formal interaction tests did not reveal statistically significant differences across consecutive vaccination groups or across key subgroups including age group, sex, and comorbidity status. Nevertheless, the observed numerical pattern is broadly consistent with previous immunological studies suggesting that repeated annual vaccination may modulate immune responses. It has been reported that individuals vaccinated repeatedly over multiple seasons often maintain higher baseline antibody levels, which may limit the magnitude of the antibody titer boost after each subsequent vaccination [25]. Additionally, repeated vaccination has been associated with reduced antibody affinity maturation, a mechanism that could potentially affect the immune system’s ability to generate optimal protective responses against antigenically drifted viruses [26–28]. However, these immunological observations do not necessarily translate into clinically meaningful reductions in VE, and our data indicate that even with longer-term consecutive vaccination, meaningful protection was still observed compared with non-consecutive vaccination. This finding aligns with a recent TND study in China, which emphasized that receiving the current season’s vaccine remains the best choice for preventing influenza in the elderly, irrespective of prior vaccination history [29]. While encouraging annual vaccination remains necessary to maintain a baseline of protection, we acknowledge that the long-term immunological interactions of repeated vaccination are complex and warrant further investigation.

In terms of clinical outcomes, vaccination appeared to provide effective protection against both moderate-to-severe influenza and influenza-associated hospitalization. These findings are clinically relevant, as reducing the severity of influenza illness in the elderly may help alleviate the burden on healthcare systems and improve patient outcomes [30]. However, due to the limited number of cases in these severity subgroups, we used all influenza-negative patients as the common control group to ensure sufficient statistical power. While this analytical approach provides more stable estimates, the results should be interpreted with caution and should not be considered conclusive. Future studies with larger sample sizes are warranted to confirm these findings, which could provide more robust evidence to inform clinical practice and public health decision-making regarding influenza vaccination strategies for the elderly population.

Limitations

This study has several limitations. First, the retrospective test-negative design may be subject to healthcare-seeking behavior bias. Although the design inherently mitigates some bias, residual confounding may persist from unmeasured factors, including healthcare-seeking patterns, prior healthcare utilization, and socioeconomic status. Second, small subgroup sample sizes (e.g., oldest-old, severe, and hospitalized cases) reduced the precision of certain VE estimates. Third, information bias may arise from the lack of household members’ vaccination data, as vaccinated contacts could provide indirect protection. Additionally, the assessment of elderly health status was limited: chronic comorbidities were self-reported and could not be verified, and we could not fully account for immunocompromised status or frailty, which may affect vaccine response and influenza susceptibility.

Conclusion

Based on influenza sentinel surveillance data from Zhejiang Province during the 2024–2025 season, this study found that influenza vaccination provided moderate overall protection against laboratory-confirmed influenza among patients aged ≥ 60 years with ILI. For consecutive vaccination, although the effectiveness of longer-term vaccination was not statistically significantly different compared with non-consecutive vaccination, given that vaccination provided meaningful protection across all groups, annual influenza vaccination and current-season vaccination are still encouraged for the elderly population. In addition, influenza vaccination provided relatively effective protection against moderate-to-severe influenza and hospitalization, highlighting its clinical value in reducing severe outcomes among older adults. Larger studies are warranted to further validate these findings.

Abbreviations

WHO

World Health Organization

VE

Vaccine Effectiveness

TND

Test-Negative Design

ILI

Influenza-like Illness

IIV3

Inactivated Trivalent Influenza Vaccine

IIV4

Inactivated Quadrivalent Influenza Vaccine

RT-PCR

Reverse Transcription Polymerase Chain Reaction

IQRs

Interquartile Ranges

OR

Odds Ratio

95% CI

95% Confidence Interval

Authors’ contributions

H.Z., J.S., X.W.: conceptualisation, study supervision, review and editing. X.L., J.J., Z.X. : data analysis, writing original draft. H.Z., X.F., T.H.: methodology design. X.L., L.H., F.Z., L.W., G.Z., Y.L., H.Y., L.Q., S.W.: data investigation and collection. J.J., Z.X., Y.C., J.P., Z.Y., S.L.: data curation, validation. L.H., T.H., L.Q., X.F., S.W.: resources provision. L.M., R.B: data visualization. All authors reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Funding

This work was supported by the National Natural Science Foundation of China (82302519); the key grants of Department of Science and Technology of Zhejiang Province (2024C03216,2025C02186); the Key Program of Health Commission of Zhejiang Province/Science Foundation of National Health Commission (WKJ-ZJ-2523); the National Key Research and Development Project by the Ministry of Science and the Technology of China (2023YFC2308705); the Zhejiang Provincial Medical and Health Science and Technology Plan (2023RC143); the Zhejiang Provincial Disease Prevention and Control Science and Technology Plan (2025JK014); the Disease Prevention and Control Innovation Team of Zhejiang Province (2026JKC-04); the Young Talent Support Program of Zhejiang Provincial Association for Science and Technology; the Young Emerging Talents in Medicine of Zhejiang Province; the Advanced Training Program for Young Elites in Public Health of the Chinese Preventive Medicine Association; the Fuyang District Social Science and Technology Development Project (2021SK025).

Data availability

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the Institutional Review Board of the Zhejiang Provincial Center for Disease Control and Prevention (No. 2024-090-01). Written informed consent was obtained from all participants or their legal guardians prior to their enrollment in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.