Abstract
Background
Respiratory infections, particularly those caused by influenza A virus (IAV), represent a major global health concern, contributing to substantial morbidity, mortality, and economic burden on healthcare systems worldwide.
Methods
We conducted a retrospective cohort study of 21,191 patient records in Chongqing, spanning a five-year surveillance period. Logistic and Poisson regression analyses were performed to characterize IAV prevalence across age-stratified, gender-specific, and clinically defined subgroups. In addition, restricted cubic spline (RCS) models were applied to examine the nonlinear effects of meteorological factors on IAV transmission.
Results
The overall IAV positivity rate was 13.79%, with marked differences across demographic subgroups. Preschool children (ages 4–6) showed the strongest association with IAV infection (OR = 6.25, 95% CI: 4.98–7.85; RR = 5.14, 95% CI: 4.13–6.41; both P < 0.001), followed by school-aged children (ages 7–18; OR = 5.60, 95% CI: 4.45–7.06; RR = 4.70, 95% CI: 3.76–5.88; both P < 0.001). Following the relaxation of COVID-19 restrictions, IAV positivity rebounded to 21.66% in 2023, with epidemic peaks observed in March–April and November–March. Four distinct epidemic waves were identified during the 2020–2024 surveillance period. RCS models revealed significant nonlinear associations between IAV prevalence and mean temperature, temperature variation, and mean relative humidity (all P for overall and nonlinear < 0.05). The highest positivity rates occurred at mean temperatures of 7.16–16.80 °C, temperature differences > 10.92 °C, and relative humidity levels of 61.91–74.00% (all P < 0.05). Furthermore, IAV infection was significantly more common in female patients compared with male patients (14.79% vs. 12.57%), and among patients with upper respiratory tract infections compared with those with lower respiratory tract infections (15.84% vs. 11.42%).
Conclusions
These findings highlight the importance of targeted public health interventions and sustained surveillance to reduce the burden of IAV-related respiratory infections and improve patient outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12967-025-07136-4.
Keywords: Influenza A virus, Climatic factors, Epidemiology, Respiratory tract infection
Introduction
Influenza A virus (IAV) remains a persistent global health threat because of its extensive geographic distribution, high mutation rate, and substantial disease burden [1]. As enveloped RNA viruses belonging to the Orthomyxoviridae family, IAV demonstrate remarkable evolutionary adaptability through antigenic drift (progressive accumulation of HA/NA gene mutations) and antigenic shift (genomic segment reassortment between human and animal strains) [2]. This genetic plasticity enables IAV to evade population immunity, resulting in annual epidemics that affect 5–20% of the global population [3]. According to estimates from the World Health Organization (WHO), approximately 10% of the global population is infected with influenza annually, with influenza-related mortality ranging from 290,000 to 650,000 deaths worldwide [4]. Historical precedents such as the 2009 H1N1 pandemic, together with ongoing zoonotic transmission at the human–livestock interface, underscore the urgent need for continuous surveillance and optimization of intervention strategies [5, 6].
IAV demonstrates tropism for the respiratory epithelium, producing a disease spectrum that ranges from self-limiting upper respiratory tract infections (URTI) to severe lower respiratory complications [7]. URTIs, characterized by nasal mucosal and pharyngeal involvement, typically present with rhinorrhea, sore throat, and low-grade fever in immunocompetent hosts, and usually resolve spontaneously within 3–5 days [7]. By contrast, lower respiratory tract infections (LRTIs) involve bronchiolar and alveolar invasion, progressing to viral pneumonia in 15–30% of hospitalized patients [8]. Notably, 20–40% of LRTI cases develop acute hypoxemic respiratory failure, which is associated with a 15-fold higher mortality compared with URTI [8]. This severity gradient is linked to differential receptor distribution: the upper airway predominantly expresses sialic acid α-2,6-galactose (Sia-α-2,6-Gal) receptors that preferentially bind human-adapted H1/H3 subtypes, whereas the lower airway is enriched in Sia-α-2,3-Gal receptors, facilitating pulmonary invasion via high-affinity binding of avian-origin H5/H7 variants [5, 6, 9]. Systemic complications, including viral myocarditis and secondary bacterial pneumonia, further exacerbate the healthcare burden by involving multiple organ systems [10].
This study aims to elucidate the epidemiological characteristics of IAV in Chongqing, China (2020–2024), and to investigate climate-mediated transmission mechanisms in order to develop region-specific prevention and control strategies. As a subtropical megacity with persistently high humidity (mean annual relative humidity 75–85%) and dense urbanization (31 million residents), Chongqing provides a unique ecological setting for examining humidity-driven viral persistence. The findings are expected to inform targeted interventions, such as the implementation of humidity-regulated public spaces and precision vaccination campaigns prior to winter surges in humidity. Ultimately, this study seeks to advance evidence-based strategies for mitigating IAV transmission in Chongqing and other humid subtropical regions worldwide.
Materials and methods
Patients
A retrospective hospital-based cohort study was conducted, enrolling 21,191 consecutive patients who presented with respiratory infection symptoms at the Chongqing Health Center for Women and Children between January 1, 2020, and December 31, 2024. All participants underwent comprehensive respiratory viral panel testing, with a primary focus on influenza A/B virus detection, to delineate the epidemiological profile and climate-associated transmission mechanisms. The study protocol was approved by the Ethics Committee of the Chongqing Health Center for Women and Children, and written informed consent was obtained from all participants prior to data collection. All procedures were performed in accordance with relevant ethical guidelines and regulations.
Subgroups
Patients were stratified into IAV-positive and IAV-negative groups according to real-time RT-PCR results. Clinically, they were classified into upper respiratory tract infection (URTI) and lower respiratory tract infection (LRTI) groups based on WHO case definitions and radiological confirmation. Patients were also divided into six age groups: 0–3 months, 4–12 months, 1–3 years, 4–6 years, 7–18 years, and > 18 years. By testing year, patients were categorized into five groups: 2020, 2021, 2022, 2023, and 2024.
Molecular detection of influenza A/B virus
Oropharyngeal swabs were collected from the bilateral tonsils and posterior pharyngeal wall using sterile flocked swabs (Copan Italia), immediately placed in 3 mL of viral transport medium, and transported at 4 °C to the central laboratory within 2 h. Aliquots of 200 μL clinical specimens, along with negative and positive controls, were processed in 1.5 mL DNase-free microcentrifuge tubes. Nucleic acid extraction was performed using the Magnetic Bead-based Pathogen Kit (Sansure Biotech Inc.) according to the manufacturer’s protocol, with elution in 60 μL of RNase-free buffer. A 5 μL aliquot of extracted nucleic acid was mixed with 45 μL of multiplex RT-PCR master mix (containing primers/probes) in 0.2 mL reaction tubes. Amplification was carried out on a CFX96 Touch™ Real-Time PCR System (Bio-Rad).
Meteorological data
Meteorological data for Chongqing from January 1, 2020, to December 31, 2024, including maximum, minimum, and mean temperatures, diurnal temperature range, and mean relative humidity, were obtained from the National Meteorological Science Data Center.
Statistical analysis
Statistical analyses were performed using STATA/MP 17.0 (StataCorp LP), and data visualization was conducted in OriginPro 2024 (OriginLab Corporation). Continuous variables (e.g., mean relative humidity, mean temperature, temperature variation) were dichotomized using receiver operating characteristic (ROC) curve analysis. Optimal cutoff values were determined by maximizing Youden’s index (J = sensitivity + specificity – 1), and variables were subsequently transformed into binary categories (0: below cutoff; 1: above cutoff).
Restricted cubic spline (RCS) analyses were applied to mean relative humidity, mean temperature, and diurnal temperature range to explore potential nonlinear associations with IAV prevalence. Based on cutoff values derived from the RCS models, additional stratified analyses were performed to assess IAV prevalence across subgroups defined by these three variables, with further stratification by age, sex, and clinical diagnosis.
For univariate analysis, Pearson’s chi-square test with Yates’ continuity correction was used for dichotomous variables, and IAV positivity rates (cases/total × 100%) were calculated across exposure categories. Multivariate analysis included: (1) logistic regression to estimate adjusted odds ratios (ORs) with 95% confidence intervals (CIs) for infection risk, and (2) Poisson regression with robust variance to calculate adjusted risk ratios (RRs). A two-sided P < 0.05 was considered statistically significant.
Results
Basic information
A total of 21,191 patients with respiratory infections were enrolled (9,567 males and 11,624 females), with ages ranging from 5 days to 57 years. Clinical diagnoses included 11,356 cases of URTI and 9,835 cases of LRTI. The overall detection rate of IAV was 13.79% (2,922/21,191). In univariate analysis, the detection rate was 17.6% higher in females than in males (14.79% vs. 12.57%, P < 0.001). Patients with URTI showed a 38.7% higher positivity rate than those with LRTI (15.84% vs. 11.42%, P < 0.001). Age-specific analysis revealed the highest prevalence among school-aged children (4–18 years: 19.30–20.25%) and the lowest among neonates (0–3 months: 2.81%). Temporal trends indicated a resurgence of IAV in 2023 (21.66% positivity, P < 0.001), coinciding with the relaxation of COVID-19 control measures (Fig. 1).
Fig. 1.
The prevalence of influenza A virus in different subgroups. PA, P value from univariate analysis. PB, P value from multivariable logistic regression analysis. PC, P value from multivariable Poisson regression analysis. a, Chi-square test, without controlling for any covariates. b, Multivariable regression analysis adjusted for age, diagnosis, and detection time. c, Multivariable regression analysis adjusted for diagnosis, gender, and detection time. d, Multivariable regression analysis adjusted for age, gender, and detection time. e, Multivariable regression analysis adjusted for age, gender, and diagnosis. URTI, Upper respiratory tract infection. LRTI, Lower respiratory tract infection. OR, Odds ratio. RR, Relative risk. CI, Confidence interval
To ensure robustness, multivariable analyses were conducted using two regression approaches: logistic regression (OR) and Poisson regression with robust variance (RR). Only results consistent across both models were considered statistically significant. In sex-stratified analysis, after adjusting for diagnostic category, age, and sampling period, no significant difference was observed between males and females. In diagnosis-stratified analysis, adjusted for sex, age, and sampling period, the positivity rate was significantly lower in LRTI compared with URTI (OR = 0.70, 95% CI: 0.64–0.76; RR = 0.75, 95% CI: 0.69–0.81; both P < 0.001). In age- and time-stratified analyses, after adjustment for relevant covariates, significant differences in IAV prevalence were also observed (details shown in Fig. 1).
Significant seasonal variation in IAV transmission dynamics was observed
To characterize seasonal patterns of IAV transmission, we examined monthly positivity rates from January 2020 to December 2024, which highlighted substantial epidemiological disruptions associated with COVID-19 containment measures. A modest epidemic peak occurred in January 2020 (positivity rate: 7.54%), followed by an almost complete absence of IAV activity from February 2020 through December 2022, attributable to strict non-pharmaceutical interventions (NPIs) that markedly reduced human mobility. An atypical summer resurgence was detected in July–August 2022 (positivity rate: 15.2%), coinciding with the temporary relaxation of mobility restrictions. This epidemic rebound was likely driven by accumulated immunity debt. After the normalization of NPIs, three additional epidemic waves were observed: March–April 2023 (spring wave: 21.55–38.52%), November 2023–March 2024 (winter–spring wave: 11.66–20.50%), and December 2024 (winter wave: 21.58%). These findings suggest a partial re-establishment of pre-pandemic seasonality, overlain by residual anomalies in viral transmission (Fig. 2A).
Fig. 2.
Monthly positivity rates of influenza A virus between 2020 and 2024. A, overall sample. B, by gender. C, by age
Demographic stratification revealed sex-specific epidemic patterns. During the March–April 2023 outbreak, positivity rates were comparable between males and females (30.11% vs. 29.97%). However, in all other epidemic waves, females consistently exhibited significantly higher infection rates (Fig. 2B). Age-stratified analysis demonstrated distinct risk profiles across major outbreak periods. Adults (> 18 years) showed peak positivity during the March–April 2023 (24.63–50.00%) and November 2023–March 2024 (14.55–30.89%) waves, whereas neonates (< 3 months) consistently exhibited the lowest infection rates (Fig. 2C).
Significant disparity in IAV detection rates was observed between URTI and LRTI
Multivariable regression analysis identified marked diagnostic-specific heterogeneity in IAV positivity during 2020–2024, with the largest difference observed in 2023 (URTI: 28.99% vs. LRTI: 15.10%; OR = 2.29, 95% CI: 2.05–2.57, P < 0.001). Across all four epidemic waves, IAV positivity remained consistently higher in URTI than in LRTI, suggesting anatomical tropism contributes to differential detection probability (Fig. 3A). Sex-stratified analysis demonstrated persistently higher IAV detection rates in females than in males across both diagnostic categories from 2022–2024, except during the December 2024 URTI wave, where positivity was comparable between sexes (male: 23.36% vs. female: 23.78%) (Fig. 3B). In contrast, among LRTI cases, sex-specific trends differed: during the March–April 2023 epidemic peak, male patients exhibited higher IAV positivity than females (Fig. 3C).
Fig. 3.
Monthly positive rate of influenza A virus based on clinical diagnosis between 2020 and 2024. A, overall sample. B, by URTI. C, by LRTI. URTI, Upper respiratory tract infection. LRTI, Lower respiratory tract infection
Mean temperature, temperature difference, mean related humility might affect the prevalence of IAV
To investigate the seasonal drivers of IAV epidemiology, meteorological data from the China National Meteorological Science Data Center (2020–2024) were analyzed for Chongqing municipality. We first applied restricted cubic splines (RCS) to explore potential nonlinear associations between climatic factors and IAV prevalence. The results demonstrated significant nonlinear effects of mean temperature, diurnal temperature range, and mean relative humidity on IAV activity (all P for overall and nonlinear < 0.05). Based on the RCS findings, mean temperature and relative humidity were categorized into three subgroups, while diurnal temperature range was divided into two subgroups (Fig. 4A,B,C).
Fig. 4.
The positivity rate of influenza A virus across subgroups defined by mean temperature, temperature variation, and mean relative humidity, with cutoff values determined from restricted cubic spline analyses. A, Mean temperature. B, Temperature variation. C, Mean relative humidity. D, Prevalence of Influenza A Virus Across Climatic Subgroups. RR, Relative risk. CI, Confidence interval. Pa, Multivariable Poisson regression analysis adjusted for age, gender and diagnosis
IAV prevalence was highest at a mean temperature of 7.16–16.80°C (1770/9626, 18.39%), but decreased by 57% when the mean temperature exceeded 16.80°C (P < 0.001). Similarly, prevalence peaked when the diurnal temperature range exceeded 10.92°C (502/2714, 18.50%), representing a 1.19-fold increase compared with ≤ 10.92°C (P < 0.001). For mean relative humidity, the highest prevalence was observed within 61.91–74.00% (1312/7490, 17.52%), whereas prevalence significantly declined when humidity exceeded 74.00% or fell below 61.91% (P < 0.001) (Fig. 4).
To ensure the robustness of these findings, non-normally distributed parameters (mean temperature, diurnal temperature range, and mean relative humidity) were further evaluated using receiver operating characteristic curve analysis with Youden index optimization to identify epidemiological thresholds. Significant increases in IAV prevalence were observed at mean temperature ≤ 19.75 °C (18.27% vs. 5.82%, P < 0.001), diurnal temperature range ≥ 10.95 °C (18.50% vs. 13.10%, P < 0.001), and relative humidity ≤ 76.5% (15.52% vs. 11.39%, P < 0.001). Multivariable regression models further confirmed the predictive validity of these thresholds. Collectively, these findings corroborate the RCS results, underscoring that fluctuations in climatic conditions significantly shape IAV transmission dynamics in Chongqing (Fig. 5).
Fig. 5.
The impact of meteorological factors on the prevalence of influenza A virus. Pa, Chi-square test, without controlling for any covariates. Pb and Pc, Multivariable regression analysis adjusted for diagnosis, gender, Age and detection time
Significant differences in the prevalence of influenza A virus were observed across sex, diagnosis, and age groups within each climatic subgroup
To further explore the influence of climatic factors on the prevalence of IAV, we stratified the population according to the RCS-derived climatic subgroups and analyzed IAV prevalence across sex, age, and clinical diagnosis. Across all subgroups of mean temperature, male patients consistently exhibited significantly lower IAV prevalence than female patients (P < 0.05). When the mean temperature was ≥ 7.16°C, patients with lower respiratory tract infections had significantly lower IAV prevalence compared with those with upper respiratory tract infections (P < 0.05), and patients aged ≥ 4 months had a markedly higher IAV positivity rate than those aged ≤ 3 months (P < 0.05) (Table 1 and supplementary Table 1).
Table 1.
The positivity rate of influenza A virus across different subgroups after grouping the mean temperature based on cutoff values obtained from the restricted cubic spline curve
| Variable | Subgroup | Positive (N) | Negative(N) | Total (N) | Positive rate (%) | Adjusted RR (95% CI) | P |
|---|---|---|---|---|---|---|---|
| Mean temperature (< 7.16 °C) | |||||||
| Gender | Female | 93 | 435 | 528 | 0.18 | Reference | Reference |
| Male | 36 | 318 | 354 | 0.10 | 0.53 (0.35–0.80) | 0.002a | |
| Diagnosis | URTI | 76 | 378 | 454 | 0.17 | Reference | Reference |
| LRTI | 53 | 375 | 428 | 0.12 | 0.74 (0.53–1.02) | 0.092b | |
| Age | 0–3 Months | 6 | 122 | 128 | 4.69 | Reference | Reference |
| 4–12 Months | 10 | 92 | 102 | 9.80 | 2.21 (0.78–6.30) | 0.191c | |
| 1–3 Years | 30 | 188 | 218 | 13.76 | 3.25 (1.31–8.03) | 0.01c | |
| 4–6 Years | 31 | 121 | 152 | 20.39 | 5.21 (2.10–12.94) | < 0.001c | |
| 7–18 Years | 10 | 98 | 108 | 9.26 | 2.08 (0.73–5.91) | 0.198c | |
| > 18 Years | 42 | 132 | 174 | 24.14 | 6.47 (2.66–15.76) | < 0.001c | |
| Mean temperature (7.16–16.80 °C) | |||||||
| Gender | Female | 1047 | 4378 | 5425 | 0.19 | Reference | Reference |
| Male | 723 | 3478 | 4201 | 0.17 | 0.87 (0.78–0.97) | 0.009a | |
| Diagnosis | URTI | 1078 | 4311 | 5389 | 0.20 | Reference | Reference |
| LRTI | 692 | 3545 | 4237 | 0.16 | 0.82 (0.75–0.89) | < 0.001b | |
| Age | 0–3 Months | 46 | 1307 | 1353 | 3.40 | Reference | Reference |
| 4–12 Months | 105 | 695 | 800 | 13.13 | 4.29 (3.00–6.14) | < 0.001c | |
| 1–3 Years | 380 | 1715 | 2095 | 18.14 | 6.30 (4.60–8.62) | < 0.001c | |
| 4–6 Years | 470 | 1469 | 1939 | 24.24 | 9.09 (6.66–12.42) | < 0.001c | |
| 7–18 Years | 418 | 1474 | 1892 | 22.09 | 8.06 (5.89–11.02) | < 0.001c | |
| > 18 Years | 351 | 1196 | 1547 | 22.69 | 8.34 (6.07–11.45) | < 0.001c | |
| Mean temperature (> 16.80 °C) | |||||||
| Gender | Female | 579 | 5092 | 5671 | 0.10 | Reference | Reference |
| Male | 444 | 4566 | 5010 | 0.09 | 0.86 (0.75–0.97) | 0.019a | |
| Diagnosis | URTI | 645 | 4867 | 5512 | 0.12 | Reference | Reference |
| LRTI | 378 | 4791 | 5169 | 0.73 | 0.62 (0.55–0.71) | < 0.001b | |
| Age | 0–3 Months | 40 | 1758 | 1798 | 2.22 | Reference | Reference |
| 4–12 Months | 54 | 1139 | 1193 | 4.53 | 2.08 (1.38–3.16) | 0.001c | |
| 1–3 Years | 219 | 2659 | 2878 | 7.61 | 3.62 (2.57–5.10) | < 0.001c | |
| 4–6 Years | 291 | 1721 | 2012 | 14.46 | 7.43 (5.31–10.41) | < 0.001c | |
| 7–18 Years | 254 | 1114 | 1368 | 18.57 | 10.02 (7.12–14.10) | < 0.001c | |
| > 18 Years | 165 | 1267 | 1432 | 11.52 | 5.72 (4.02–8.15) | < 0.001c | |
URTI Upper respiratory tract infection, LRTI Lower respiratory tract infection, RR Relative risk, CI Confidence interval.
a, Multivariable Poisson regression analysis adjusted for age and diagnosis. b, Multivariable Poisson regression analysis adjusted for age and gender. c, Multivariable Poisson regression analysis adjusted for diagnosis and gender.
In diurnal temperature range subgroups, patients with lower respiratory tract infections exhibited consistently lower IAV prevalence than those with upper respiratory tract infections (P < 0.05). When the temperature variation was ≤ 10.92°C, male patients had significantly lower IAV prevalence than female patients (P < 0.05), and patients aged ≥ 1 year showed significantly higher positivity rates than those aged ≤ 3 months (P < 0.05) (Table 2 and supplementary Table 2).
Table 2.
The positivity rate of influenza A virus across different subgroups after grouping the diurnal temperature range based on cutoff values obtained from the restricted cubic spline curve
| Variable | Subgroup | Positive (N) | Negative(N) | Total (N) | Positive rate (%) | Adjusted RR (95% CI) | P |
|---|---|---|---|---|---|---|---|
| Diurnal temperature range (< = 10.92°C) | |||||||
| Gender | Female | 1449 | 8693 | 10,142 | 14.29 | Reference | Reference |
| Male | 971 | 7364 | 8335 | 11.65 | 0.79 (0.73–0.86) | < 0.001a | |
| Diagnosis | URTI | 1468 | 8301 | 9769 | 15.03 | Reference | Reference |
| LRTI | 952 | 7756 | 8708 | 10.93 | 0.69 (0.64–0.76) | < 0.001b | |
| Age | 0–3 Months | 73 | 2832 | 2905 | 2.51 | Reference | Reference |
| 4–12 Months | 155 | 1713 | 1868 | 8.30 | 3.51 (2.64–4.67) | < 0.001c | |
| 1–3 Years | 538 | 3994 | 4532 | 11.87 | 5.23 (4.07–6.70) | < 0.001c | |
| 4–6 Years | 636 | 2934 | 3570 | 17.82 | 8.41 (6.57–10.77) | < 0.001c | |
| 7–18 Years | 517 | 2294 | 2811 | 18.39 | 8.74 (6.80–11.24) | < 0.001c | |
| > 18 Years | 501 | 2290 | 2791 | 17.95 | 8.49 (6.60–10.92) | < 0.001c | |
| Diurnal temperature range (> 10.92°C) | |||||||
| Gender | Female | 270 | 1212 | 1482 | 18.22 | Reference | Reference |
| Male | 232 | 1000 | 1232 | 18.83 | 1.04 (0.86–1.27) | 0.691a | |
| Diagnosis | URTI | 331 | 1256 | 1587 | 20.86 | Reference | Reference |
| LRTI | 171 | 956 | 1127 | 15.17 | 0.68 (0.55–0.83) | < 0.001b | |
| Age | 0–3 Months | 19 | 355 | 374 | 5.08 | Reference | Reference |
| 4–12 Months | 14 | 213 | 227 | 6.17 | 1.23 (0.60–2.50) | 0.583c | |
| 1–3 Years | 91 | 569 | 660 | 13.79 | 2.99 (1.79–4.99) | < 0.001c | |
| 4–6 Years | 156 | 378 | 534 | 29.21 | 7.71 (4.69–12.69) | < 0.001c | |
| 7–18 Years | 165 | 392 | 557 | 29.62 | 7.87 (4.79–12.92) | < 0.001c | |
| > 18 Years | 57 | 305 | 362 | 15.75 | 3.49 (2.03–6.00) | < 0.001c | |
URTI Upper respiratory tract infection, LRTI Lower respiratory tract infection, RR Relative risk, CI Confidence interval.
a, Multivariable Poisson regression analysis adjusted for age and diagnosis. b, Multivariable Poisson regression analysis adjusted for age and gender. c, Multivariable Poisson regression analysis adjusted for diagnosis and gender
In mean relative humidity subgroups, patients with lower respiratory tract infections also demonstrated significantly lower IAV prevalence than those with upper respiratory tract infections (P < 0.05), and patients aged ≥ 1 year consistently showed higher positivity rates than those aged ≤ 3 months (P < 0.05). Notably, within the relative humidity range of 61.91–74.00%, where IAV prevalence was highest, no significant sex differences were observed. However, in all other humidity subgroups, male patients had significantly lower IAV prevalence than female patients (P < 0.05) (Table 3 and supplementary Table 3).
Table 3.
The positivity rate of influenza A virus across different subgroups after grouping the mean relative humidity based on cutoff values obtained from the restricted cubic spline curve
| Variable | Subgroup | Positive (N) | Negative(N) | Total (N) | Positive rate (%) | Adjusted RR (95% CI) | P |
|---|---|---|---|---|---|---|---|
| Mean relative humidity (< 61.91%) | |||||||
| Gender | Female | 163 | 1328 | 1491 | 0.11 | Reference | Reference |
| Male | 111 | 1174 | 1285 | 0.09 | 0.77 (0.60–0.99) | 0.048a | |
| Diagnosis | URTI | 178 | 1363 | 1541 | 0.12 | Reference | Reference |
| LRTI | 96 | 1139 | 1235 | 0.08 | 0.65 (0.50–0.84) | 0.001b | |
| Age | 0–3 Months | 17 | 491 | 508 | 0.03 | Reference | Reference |
| 4–12 Months | 21 | 332 | 353 | 0.06 | 1.83 (0.95–3.52) | 0.090c | |
| 1–3 Years | 68 | 677 | 745 | 0.09 | 2.90 (1.68–5.00) | < 0.001c | |
| 4–6 Years | 68 | 383 | 451 | 0.15 | 5.13 (2.96–8.87) | < 0.001c | |
| 7–18 Years | 41 | 230 | 271 | 0.15 | 5.15 (2.86–9.26) | < 0.001c | |
| > 18 Years | 59 | 389 | 448 | 0.13 | 4.38 (2.51–7.64) | < 0.001c | |
| Mean relative humidity (61.91–74.00%) | |||||||
| Gender | Female | 744 | 3384 | 4128 | 0.18 | Reference | Reference |
| Male | 568 | 2794 | 3362 | 0.17 | 0.93 (0.82–1.04) | 0.210a | |
| Diagnosis | URTI | 826 | 3284 | 4110 | 0.20 | Reference | Reference |
| LRTI | 486 | 2894 | 3380 | 0.14 | 0.67 (0.59–0.76) | < 0.001b | |
| Age | 0–3 Months | 36 | 980 | 1016 | 0.04 | Reference | Reference |
| 4–12 Months | 69 | 651 | 720 | 0.10 | 2.89 (1.91–4.37) | < 0.001c | |
| 1–3 Years | 269 | 1530 | 1799 | 0.15 | 4.79 (3.35–6.84) | < 0.001c | |
| 4–6 Years | 392 | 1155 | 1547 | 0.25 | 9.24 (6.50–13.13) | < 0.001c | |
| 7–18 Years | 347 | 1008 | 1355 | 0.26 | 9.37 (6.58–13.36) | < 0.001c | |
| > 18 Years | 199 | 854 | 1053 | 0.19 | 6.34 (4.40–9.15) | < 0.001c | |
| Mean relative humidity (> 74.00%) | |||||||
| Gender | Female | 812 | 5185 | 5997 | 0.14 | Reference | Reference |
| Male | 524 | 4383 | 4907 | 0.11 | 0.76 (0.68–0.86) | < 0.001a | |
| Diagnosis | URTI | 795 | 4906 | 5701 | 0.14 | Reference | Reference |
| LRTI | 541 | 4662 | 5203 | 0.10 | 0.72 (0.64–0.80) | < 0.001b | |
| Age | 0–3 Months | 39 | 1712 | 1751 | 0.02 | Reference | Reference |
| 4–12 Months | 79 | 935 | 1014 | 0.08 | 3.71 (2.51–5.49) | < 0.001c | |
| 1–3 Years | 292 | 2348 | 2640 | 0.11 | 5.46 (3.89–7.67) | < 0.001c | |
| 4–6 Years | 332 | 1773 | 2105 | 0.16 | 8.22 (5.86–11.53) | < 0.001c | |
| 7–18 Years | 294 | 1448 | 1742 | 0.17 | 8.91 (6.34–12.54) | < 0.001c | |
| > 18 Years | 300 | 1352 | 1652 | 0.18 | 9.74 (6.93–13.70) | < 0.001c | |
URTI Upper respiratory tract infection. LRTI Lower respiratory tract infection. RR Relative risk. CI Confidence interval.
a, Multivariable Poisson regression analysis adjusted for age and diagnosis. b, Multivariable Poisson regression analysis adjusted for age and gender. c, Multivariable Poisson regression analysis adjusted for diagnosis and gender
Mixed infection of influenza A virus and influenza B virus will not increase the incidence rate of LRTI
To investigate potential variations IAV infection patterns across demographic and clinical subgroups, we conducted a stratified analysis of mono- versus co- infection status. The study cohort comprised 2,922 IAV-positive patients (2020–2024), with 2,899 (99.21%) mono-infections and 23 (0.79%) IAV/IBV coinfections confirmed through multiplex PCR assays. Subgroup analysis revealed that the incidence of IAV/IBV coinfections was higher in female patients, LRTI, and patients with 0–3 months, but there was no statistical difference (Fig. 6).
Fig. 6.
The potential influence of IAV mono-infections and IAV/IBV coinfections across demographic and clinical subgroups. A, by gender. B, by diagnosis. C, by age. D, detection time. URTI, Upper respiratory tract infection. LRTI, Lower respiratory tract infection
Discussion
Understanding IAV infections remains critical due to their substantial impact on global public health [11, 12]. As a major respiratory pathogen, IAV causes seasonal epidemics and occasional pandemics, resulting in significant morbidity and mortality worldwide, particularly among vulnerable populations such as the elderly and individuals with pre-existing conditions [13, 14]. Challenges persist despite advances in surveillance and antiviral strategies, including the emergence of drug-resistant strains and the need for timely, accurate diagnostics [15]. In this context, the present study examines the epidemiology of IAV infections and investigates demographic and clinical factors that influence susceptibility and detection, providing insights to inform targeted prevention and control measures.
In the present study, we observed that the positivity rate of IAV was significantly higher in patients with upper respiratory tract infections (15.84%) compared to those with lower respiratory tract infections (11.42%). This pattern aligns with the known tropism of IAV, which preferentially infects epithelial cells of the upper airway, where sialic acid receptors with α2,6 linkages are highly expressed, facilitating viral attachment and entry [5]. Upon infection, IAV rapidly replicates in the ciliated and goblet cells of the nasal and tracheal epithelium, triggering local inflammation and mucus hypersecretion, which contribute to higher viral loads in the upper respiratory tract [7]. In contrast, the lower respiratory tract expresses a greater proportion of α2,3-linked sialic acid receptors, which are less efficiently targeted by human-adapted IAV strains, potentially explaining the comparatively lower positivity rates in lower respiratory tract infections [16]. Furthermore, the host innate immune responses in the lower airway, including alveolar macrophages and type I interferon signaling, may provide additional barriers that limit viral replication [8]. These findings suggest that the higher prevalence of IAV in URI cases reflects both the viral receptor specificity and the differential susceptibility of upper versus lower airway epithelium, highlighting the importance of anatomical and molecular determinants in shaping infection patterns.
Beyond infection site–related differences, our findings indicate that susceptibility and prevalence of IAV vary significantly across age groups. Previous studies suggest that such differences are influenced not only by environmental exposure or host immune status but also by interactions between the virus and host mitochondria [17, 18]. Mechanistic research has demonstrated that IAV remodels mitochondrial function during replication and dissemination [18]. Viral proteins can interfere with mitochondrial antiviral-signaling protein (MAVS), dampening type I interferon production and early antiviral responses [2]. Infection also promotes mitochondrial reactive oxygen species (mtROS) generation, amplifying inflammation and tissue damage [3]. Additionally, IAV manipulates mitochondrial dynamics to reshape the cellular metabolic environment, creating favorable conditions for viral replication while weakening host defences [1, 9]. In light of our results, we speculate that younger children (< 4 years) may have an underdeveloped immune system with less robust mitochondrial signaling, which could partially limit viral replication and pathogenesis [19]. By contrast, in older children and adults (≥ 4 years), a more mature immune system and heightened mitochondrial activity in antiviral defense may provide a more fertile ground for the virus to exploit mitochondrial pathways for immune evasion and inflammatory amplificatio n [18]. This could explain the higher positivity rates and risk of symptomatic infection observed in these groups. Thus, the age-stratified differences in IAV prevalence observed in our study may reflect not only epidemiological trends but also the underlying interplay between mitochondrial development and virus–host interactions.
Similarly, we observed a higher IAV positivity rate in female patients. This sex-specific disparity may be influenced by both biological and immunological factors. Sex hormones, particularly estrogens, can modulate innate and adaptive immune responses, potentially affecting viral replication and host susceptibility [20]. Sex-dependent differences in mitochondrial function and interferon signaling may further influence virus–host interactions and early viral clearance [18]. Nonetheless, the precise mechanisms underlying sex-specific differences remain to be elucidated, warranting further investigation.
Seasonal and geographic variability in IAV prevalence is also evident [21]. In temperate regions such as North America and Europe, annual IAV attack rates during peak seasons typically range from 10–25%, whereas tropical and subtropical areas display more prolonged but lower-intensity transmission patterns [22]. Notably, post-COVID-19 surveillance data indicate a global resurgence of IAV activity from 2022 to 2023 [23]. Despite these global insights, knowledge regarding transmission dynamics in Southwest China, particularly in Chongqing, a megacity with a subtropical monsoon climate and over 32 million residents, remains limited. Previous studies largely focused on pre-2020 epidemiology, leaving post-pandemic patterns largely uncharacterized. In our retrospective cohort integrating virological surveillance and meteorological datasets (2020–2024), we identified distinct climatic conditions associated with peak IAV positivity in Chongqing: mean temperature between 7.16–16.80 °C, mean relative humidity of 61.91–74.00%, and diurnal temperature range exceeding 10.92 °C. Positivity rates decreased significantly when mean temperature or relative humidity fell outside these ranges, indicating that environmental drivers, including temperature, humidity, and temperature variability, act as important modulators of viral transmission.
Mechanistically, climatic factors influence IAV spread through multiple pathways [24]. First, viral stability and survival are highly sensitive to temperature and humidity; influenza virions persist longer under cool, moderately humid conditions, whereas extreme heat or very high/low humidity accelerates viral inactivation [4]. Second, aerosol generation and droplet dynamics are affected by environmental conditions; lower temperatures and intermediate humidity favor the formation and airborne stability of respiratory droplets, enhancing exposure of the upper respiratory tract [22]. Third, host susceptibility may be modulated by climate-induced changes in airway physiology and mucosal immunity [25]. Low humidity can impair mucociliary clearance, reduce antiviral peptide activity, and promote viral deposition in the upper respiratory tract, while temperature fluctuations can induce stress responses that transiently compromise innate defences [25]. These insights highlight the complex interplay between environmental drivers, viral dynamics, and host factors in shaping IAV seasonality and outbreak intensity, emphasizing the importance of incorporating climatic variables into predictive models and public health strategies.
In this study, we investigated the temporal fluctuations of IAV epidemics during 2020–2024. Our longitudinal surveillance identified four distinct epidemiological peaks over the five-year period. Three of these peaks clustered within the winter-spring seasons. These seasons are characterized by lower temperatures, cold and arid climatic conditions that enhance viral persistence and transmissibility [26]. IAV predominantly spreads via respiratory droplets generated through coughing and sneezing [27]. During winter, prolonged indoor congregate settings elevate interpersonal contact frequency, thereby amplifying transmission opportunities [10]. Additionally, cold-induced immunosuppression, attributable to thermoregulatory stress and reduced sunlight exposure, may heighten host susceptibility [10]. These findings corroborate our earlier conclusion.
This study has several limitations that warrant consideration. First, although our data were derived from a large sample at the largest maternal and child health hospital in Chongqing and covered the citywide population, the single-center design may limit the generalizability of the findings. Second, because the circulation of influenza A viruses (IAVs) is influenced by both climatic and demographic differences, the observed patterns in Chongqing may exhibit regional specificity. Thus, future multicenter studies, ideally spanning multiple provinces or even international sites, are warranted to validate the broader applicability of our observations. Third, the lack of genetic subtyping of IAV in this study precluded us from distinguishing the epidemiological features and transmission dynamics of specific subtypes such as H1N1 or H3N2 under varying climatic conditions, which calls for more detailed molecular investigations. In addition, as our cohort primarily included outpatients, key host-related variables (e.g., baseline immune status, chronic comorbidities, vaccination history) were not available, which restricts a comprehensive assessment of host–climate interactions. Finally, at the mechanistic level, evidence directly linking climatic factors to IAV transmission in human populations remains scarce. While experimental studies have suggested that temperature and humidity can influence aerosol stability, viral persistence, and potentially host susceptibility, such findings are largely indirect and require further validation in real-world settings. These caveats underscore the need for integrating epidemiological surveillance with molecular and mechanistic research to strengthen causal inferences regarding the role of climate in shaping IAV epidemics.
In summary, persistent high detection rates of IAV were observed in our large sample size surveillance at a hospital in Chongqing. IAV positivity was significantly associated with preschool children (4–6 years), school-aged children (7–18 years), upper respiratory tract infection, and seasons meeting meteorological thresholds (mean temperature: 7.16–16.80 °C, mean relative humidity: 61.91–74.00%, and diurnal temperature range: 10.92 °C). Local public health authorities should establish early warning systems for seasonal IAV infection risks and prioritize targeted interventions for high-risk population groups. The transient suppression of IAV activity post-COVID-19, followed by a sharp rebound, emphasizes the importance of sustained protective behaviors. Influenza A/B co-infection did not significantly modify risk disparities across gender, age, or clinical diagnosis. Future research should expand on these findings by addressing several key gaps. Multicenter and geographically diverse surveillance is needed to evaluate whether the patterns observed in Chongqing are generalizable to other regions. Genetic subtyping of circulating IAV strains would clarify subtype-specific transmission dynamics, particularly under varying climatic conditions. Integration of comprehensive host data, including vaccination history, immune status, and comorbidities, could elucidate how host factors interact with environmental drivers to influence susceptibility. Mechanistic studies are warranted to investigate how temperature, humidity, and diurnal variation affect viral stability, aerosol transmission, and host immune responses, thereby strengthening causal inference. Finally, combining real-time meteorological monitoring with predictive modeling may improve early warning systems and guide targeted public health interventions.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgments
Not applicable.
Abbreviations
- IAV
Influenza A virus
- URTI
Upper respiratory tract infection
- LRTI
Lower respiratory tract infection
- NPIs
Non-pharmaceutical interventions
- ROC
Receiver operating characteristic
- RCS
Restricted cubic spline
Author contributions
Shu Li conceptualized the study, visualized the data, developed the methodology used, applied the software, performed the statistical analysis, wrote the original draft, and reviewed and edited the manuscript. Kewei Xing applied the software and developed the methodology used. Yan Su visualized the data. Xinyuan Zhang collected data and reviewed the manuscript. Chunli Li conceptualized the study, wrote, reviewed and edited the manuscript. Shu Li and Chunli Li confirm the authenticity of all the raw data. All authors have read and approved the final version of the manuscript.
Funding
This work was supported by the Natural Science Foundation of Chongqing (CSTB2024NSCQ-MSX0249), the National Natural Science Foundation of China (82402014), and the Chongqing medical scientific research project (Joint project of Chongqing Health Commission and Science and Technology Bureau) (2024QNXM060).
Availability of data and materials
All raw data was deposited on Zenodo [28].
Code availability
No custom codes were used in this study.
Declarations
Ethics approval and consent to participate
The present study was approved by the Ethics Association of Chongqing Health Center for Women and Children (approval no. 2024 ethics department 052).
Consent for publication
All authors have approved the manuscript for submission.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Shu Li and Kewei Xing have contributed equally to this work and share first authorship.
Chunli Li and Xinyuan Zhang are co–corresponding authors who have contributed equally to this work.
Change history
11/5/2025
Authorship has been corrected.
Contributor Information
Xinyuan Zhang, Email: cqfy_zhangxy@126.com.
Chunli Li, Email: lcl518023@126.com.
References
- 1.Kim J, Yuan Y, Agaronyan K, Zhao A, Wang VD, Gau D, et al. Damage sensing through TLR9 regulates inflammatory and antiviral responses during influenza infection. Mucosal Immunol. 2025;18(3):537–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sun L, Feng H, Misumi I, Shirasaki T, Hensley L, González-López O, et al. Viral protease cleavage of MAVS in genetically modified mice with hepatitis A virus infection. J Hepatol. 2023;78(2):271–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chen F, Matsuda A, Sporn PHS, Casalino-Matsuda SM. Hypercapnia increases influenza A virus infection of bronchial epithelial cells by augmenting cellular cholesterol via mTOR and Akt. Int J Mol Sci. 2025;26(9):4133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Deng X, Liu J, Wang M, Chen N, Hao F, Yang J, et al. Changes in influenza-associated excess mortality in China between 2012–2019 and 2020–2021: a population-based statistical modelling study. Infect Dis Poverty. 2025;14(1):52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Liu M, Bakker AS, Narimatsu Y, van Kuppeveld FJM, Clausen H, de Haan CAM, et al. H3N2 influenza A virus gradually adapts to human-type receptor binding and entry specificity after the start of the 1968 pandemic. Proc Natl Acad Sci U S A. 2023;120(31):e2304992120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Imai M, Ueki H, Ito M, Iwatsuki-Horimoto K, Kiso M, Biswas A, et al. Highly pathogenic avian H5N1 influenza a virus replication in ex vivo cultures of bovine mammary gland and teat tissues. Emerg Microbes Infect. 2025;14(1):2450029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Roach SN, Shepherd FK, Mickelson CK, Fiege JK, Thielen BK, Pross LM, et al. Tropism for ciliated cells is the dominant driver of influenza viral burst size in the human airway. Proc Natl Acad Sci U S A. 2024;121(31):e2320303121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Beppu AK, Zhao J, Yao C, Carraro G, Israely E, Coelho AL, et al. Epithelial plasticity and innate immune activation promote lung tissue remodeling following respiratory viral infection. Nat Commun. 2023;14(1):5814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chang X, Zhou S, Liu J, Wang Y, Guan X, Wu Q, et al. Zishen Tongyang Huoxue decoction (TYHX) alleviates sinoatrial node cell ischemia/reperfusion injury by directing mitochondrial quality control via the VDAC1-β-tubulin signaling axis. J Ethnopharmacol. 2024;320:117371. [DOI] [PubMed] [Google Scholar]
- 10.Chang M, Shi S, Jin Y, Wang G, Peng R, An J, et al. Epidemiology and genetic evolutionary analysis of influenza virus among children in Hainan Island China 2021–2023. Pathogens. 2025;14(2):142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yang J, Shi J, Xu W, Mu J, Chang Y, Zhang X, et al. Effects of modified Xuanbai Chengqi decoction against secondary streptococcus pneumoniae infection following influenza virus infection: a multi-omics analysis. J Ethnopharmacol. 2025. 10.1016/j.jep.2025.120458. [DOI] [PubMed] [Google Scholar]
- 12.Martins RP, Marc D, Germon P, Trapp S, Caballero-Posadas I. Influenza A virus in dairy cattle: infection biology and potential mammary gland-targeted vaccines. NPJ Vaccines. 2025;10(1):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Thomas MN, Janzen GM, Markin A, Sharma A, Hewitt K, Li G, et al. Active surveillance for influenza A virus in swine reveals within-farm reassortment and cocirculation of distinct subtypes and genetic clades. Vet Microbiol. 2025;309:110681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Brüssow H. The arrival of highly pathogenic avian influenza viruses in North America, ensuing epizootics in poultry and dairy farms and difficulties in scientific naming. Microb Biotechnol. 2024;17(12):e70062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Basu Thakur P, Bullock HA, Stevens J, Kumar A, Maines TR, Belser JA. Heterogeneity across mammalian- and avian-origin A(H1N1) influenza viruses influences viral infectivity following incubation with host bacteria from the human respiratory tract. Microbiol Spectr. 2025. 10.1128/spectrum.00977-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jiang H, Zhang Z. Immune response in influenza virus infection and modulation of immune injury by viral neuraminidase. Virol J. 2023;20(1):193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Li Y, Yu J, Li R, Zhou H, Chang X. New insights into the role of mitochondrial metabolic dysregulation and immune infiltration in septic cardiomyopathy by integrated bioinformatics analysis and experimental validation. Cell Mol Biol Lett. 2024;29(1):21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fuchs AL, Singh B, Jetmore JW, Warren EB, Banerjee P, Marin Franco JL, et al. Mitochondrial dysfunction enhances influenza pathogenesis by up-regulating de novo sialic acid biosynthesis. Sci Adv. 2025;11(27):eadu3739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Juhn YJ, Wi CI, Ryu E, King KS, Sohn S, Sagheb E, et al. Artificial intelligence biomarker detects high-risk childhood asthma subgroup for respiratory infections and exacerbations. J Allergy Clin Immunol. 2025;S0091–6749(25):00861–9. [Google Scholar]
- 20.Cervantes O, Cruz Talavera I, Every E, Coler B, Li M, Li A, et al. Role of hormones in the pregnancy and sex-specific outcomes to infections with respiratory viruses. Immunol Rev. 2022;308(1):123–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kakoullis L, Steffen R, Osterhaus A, Goeijenbier M, Rao SR, Koiso S, et al. Influenza: seasonality and travel-related considerations. J Travel Med. 2023;30(5):taad102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li D, Liu Y, Qi L. Effect of demographic and seasonal variability on an influenza epidemic in a metapopulation model. Math Biosci. 2025;386:109465. [DOI] [PubMed] [Google Scholar]
- 23.Lv T, Chen J, Li H, Chen X, Zhang N, Ma C, et al. Influenza A virus continues to circulate among children in Linyi, northern China, after the relaxation of COVID-19 control measures. Sci Rep. 2024;14(1):30164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Couto P, Campbell H, Li Y, Rondy M, Leite J, Rodriguez A, et al. Characterisation of the respiratory syncytial virus seasonality and its environmental factors in the Americas-a multi-country observational study using routine surveillance networks. Lancet Regional Health Am. 2025;48:101166. [Google Scholar]
- 25.Zheng C, Jiang X, Yin Y, Dai Q, Tang S, Hu J, et al. Exploration of the impact of air pollutants on the influenza epidemic after the emergence of COVID-19: based on Jiangsu province, China (2020–2024). Front Public Health. 2025;13:1555430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhang L, Lin J, Zheng J, Niu Y, Meng F, Liu S, et al. Temporal shifts in climate-influenza dynamics: a multi-subtype analysis in Southern China spanning the COVID-19 Era. Transbound Emerg Dis. 2025;2025:5581162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Huang X, Zhang Y, Chen J, Yang Z, Chen M, Xue Z, et al. A multiplexed three-channel detection system for rapid home-based diagnosis of respiratory viruses. Talanta. 2025;297(Pt B):128716. [DOI] [PubMed] [Google Scholar]
- 28.Li, S. Epidemic rate of influenza A virus in Chongqing, China from 2020 to 2024. Zenodo. 2025.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All raw data was deposited on Zenodo [28].
No custom codes were used in this study.