Spatio-temporal analysis of influenza in old-new economic transition areas of coastal mega-cities: evidence from China

Abstract

Background

Long-term spatio-temporal analyses of influenza in coastal mega-cities remain limited. This study explores influenza dynamics in Tianjin, China (2010–2023) to inform prevention strategies.

Methods

The data was based on case data from Tianjin (2010–2023) and the Tianjin Statistical Yearbook. Temporal trends, spatial auto-correlation (global and local Moran’s I), and spatio-temporal clusters (scan statistics) were assessed.

Results

From 2010 to 2023, Tianjin reported a cumulative total of 195,426 influenza cases. The eastern region displayed the top three prevalence indices among the sixteen districts. Spatial auto-correlation analyses indicated that the evolution of influenza incidence rates in Tianjin follows a pattern of being “high along the eastern coast”. The analysis of spatio-temporal scanning identified the most significant cluster in eastern Tianjin, occurring from December 2017 to May 2019. This cluster covered four districts with a relative risk (RR) of 7.73 and log likelihood ratio (LLR) of 6597.49 (P < 0.001). Additionally, a similar cluster emerged in the eastern region of Tianjin from December 2019 to December 2023, covering three districts (RR: 27.41; LLR: 74011.43; P < 0.001).

Conclusions

Mega-city influenza prevention prioritizes high-economic-activity zones. Coastal mega-cities target influenza spread in old-new economic transition areas.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13690-026-01873-8.

Text Box 1. Contributions to the literature
• The influenza incidence rates in Tianjin from 2010 to 2023 displayed notable spatial and temporal aggregation patterns.
• Influenza prevention and control in mega-cities focus on economic areas.
• Coastal mega-cities would focus on the transmission of influenza in the transitional zones between older and newer economic centers.
• This study provides a theoretical framework for influenza prevention and control in coastal mega-cities. The spatio-temporal patterns of influenza in these areas necessitate close attention and proactive measures to mitigate the spread of influenza.

Introduction

Influenza is an acute viral respiratory disease caused by infection of the respiratory tract with influenza viruses that circulate among people worldwide [1]. This virus spreads rapidly and can trigger local outbreaks or widespread epidemics [2]. In China, influenza currently constitutes nearly half of all reported respiratory infectious disease cases [3, 4]. According to the World Health Organization (WHO), seasonal influenza epidemics result in an estimated 650,000 global deaths annually — equivalent to approximately one death every 48s [5]. Influenza outbreaks led to heavy health and economic burdens for people around the world, which was considered a serious global public health issue [6]. China had a high burden of influenza infections. Studies reveal a significant annual influenza mortality burden in China, with an estimated 88,000 deaths occurring on average. This national toll equates to about 13.6% of all influenza-related deaths globally [7]. Moreover, Tianjin is a coastal mega-city in China with a resident population of more than 10 million. Due to its large population, Tianjin has an increased risk of influenza. Bordering the Bohai Sea to the east, Tianjin is an important outlet to the sea for neighboring landlocked countries and the largest port city in northern China. The economic level and population size of Tianjin [8] are already at the same level as in many other countries, such as Hungary [9], Sweden [10], and Greece [11]. Given this global relevance, elucidating the spatial and temporal dynamics of influenza incidence rates in Tianjin is critical to informing effective public health strategies. Such analysis not only advances our understanding of influenza transmission within coastal megacities but also provides insights into its potential for cross-border spread.

Spatial epidemiology, propelled by advances in geographic information systems (GIS) and spatial analysis technologies, has become a widespread adopted in infectious diseases research [12]. By characterizing the spatial distribution and dynamics of diseases, this discipline provides critical insights for guiding prevention strategies, improving public health outcomes, and optimizing resource allocation [13]. For instance, GIS techniques have been leveraged to map and analyze the spatio-temporal distribution of influenza A (H7N9) cases in China, pinpointing hotspots and informing outbreak response [14]. While the core focus of spatial epidemiology is to investigate spatio-temporal patterns of disease occurrence [10–12], there remains a relative scarcity of studies applying long-term spatio-temporal aggregation analysis to influenza transmission in coastal mega-cities, leaving its epidemic trends uncertain. Therefore, this study aims to analyze the spatio-temporal evolution of influenza incidence rates in Tianjin, a coastal mega-city of China, to generate localized, evidence-based insights that can strengthen influenza prevention and control measures.

Methods

Study site

This study area, Tianjin, is situated within the geographical coordinates of 116°43’–118°04’ E and 38°34’–40°15’ N. It is located in northern China. With its eastern border along the Bohai Sea, Tianjin is an important outlet to the sea for neighboring landlocked countries and the largest port city in northern China. Tianjin has sixteen districts (Heping, Hedong, Hexi, Hebei, Nankai, Hongqiao, Beichen, Jinghai, Jinnan, Jizhou, Wuqing, Baodi, Binhai, Dongli, Xiqing, and Ninghe districts), spans a land area of 11966.45 km², and has a permanent resident population of 13.63 million until 2022. In 2023, Tianjin realized a gross regional product of 2,293.01 billion dollars, an increase of 4.3% over the previous year. Consequently, study on influenza conducted in Tianjin can offer a valuable reference for shaping prevention and control strategies not only in mega-cities but also at the national level worldwide.

Data source

Influenza case data for Tianjin spanning 2010 to 2023 were collected from China’s National Notifiable Infectious Disease Reporting System (NIDRIS). The confirmation of all cases was based on the official diagnostic criteria established by National Health Commission of the people’s Republic of China. This confirmation required each case meet the clinical and laboratory benchmarks [15]. Population and regional data were obtained from the Tianjin Statistical Yearbook (https://stats.tj.gov.cn/tjsj_52032/tjnj/). The vector map used for Tianjin was sourced from the Chinese National Fundamental Geographic Information System (http://www.ngcc.cn/ngcc/).

Statistical analysis

Descriptive analysis

An analysis of influenza epidemic in Tianjin from 2010 to 2023 was performed, encompassing reported cases, incidence rates, time of onset, demographic (gender, age, and occupation), and geographic distribution. Furthermore, incidence rates for each district were calculated based on annual population data and case numbers, with findings visualized through statistical charts and maps.

Epidemic temporal indices analysis

This study analyzed data spanning January 2010 to December 2023 (168 consecutive months), which served as the basis for calculating all time-based epidemic indices. To evaluate the extent and impact of influenza risk, three temporal indices were applied, measuring (a) the frequency of disease occurrence, (b) the duration of epidemics, and (c) the intensity of case concentration during outbreaks, with each addressing a distinct dimension of the temporal risk profile [16, 17].

Within the context of an outbreak, disease frequency is defined as the probability of observing at least one laboratory-confirmed infection during any given month over the study period. This index, termed the frequency (), is defined as follows:

Where TM denotes the complete count of months spanning the study period. CM refers to the total number of months within that the study period during which at least one laboratory-confirmed case was recorded.

The duration index () serves as a measure of outbreak persistence, quantified as the mean number of months with ongoing case reports across the study period. This measure is expressed as follows:

where CM retains its earlier definition. An epidemic wave in this study refers to a consecutive occurrence of cases (at least two months). PV represents the entire amount of epidemic waves that occurred throughout the study period. This measure is used to evaluate how well public health initiatives mitigate and prevent disease. Conversely, a higher duration index reflects greater persistence of the outbreak, indicating challenges in its termination.

Intensity is defined as the probable scale of an outbreak. It refers specifically to the estimated severity within an epidemic wave when case counts exceed one. The intensity index ( ) is defined as:

where IR refers to the crude incidence rate within the study period. The crude influenza incidence rate is defined as the total number of laboratory-confirmed influenza cases per 100,000 population. This index assesses the temporal clustering of cases. A higher intensity index indicates that cases are concentrated in fewer epidemic waves (i.e., outbreaks are more acute when they occur), whereas a lower value suggests cases are more evenly distributed across a larger number of epidemic waves.

Spatial auto-correlation analysis

Spatial auto-correlation statistics are widely employed to study the distribution patterns and structural characteristics of diseases, as they can assess spatial dependence or auto-correlation within geographic datasets. This analytical approach encompasses both global and local spatial auto-correlation analysis. Global spatial auto-correlation serves to quantify the general level of spatial dependence present across an entire dataset. Local spatial auto-correlation employs local indicators of spatial association (LISA) to evaluate how individual locations influence overall spatial statistics and to identify the positions and categories of clustered patterns. The global Moran’s I index was applied to examine the presence of general spatial clustering in influenza incidence rates throughout Tianjin (18). The formula of Moran’s I is as follows:

LISA is calculated as follows (18):

Where W_ij is the matrix of spatial weights between district i and j, n is the amount of districts, x_i and x_j are the influenza incidence rates for districts i and j, respectively, and is the overall average incidence rate for all districts. The Moran’s I statistic for global auto-correlation falls between − 1 and 1. Based on statistical significance, the sign of Moran’s I reveals the spatial association: a negative value demonstrates a negative correlation, whereas a positive value is inferred to indicate a positive correlation. The strength of the spatial association corresponds directly to the magnitude of Moran’s I, where a greater numerical value indicates a stronger relationship. Based on LISA mapping, four spatial clustering patterns were identified: High-High (HH) clusters, which are characterized by regions of high incidence rates adjacent to other high-incidence regions, commonly termed hotspots; Low-Low (LL) clusters, defined as regions with low incidence rates neighboring other low-incidence regions, known as coldspots; High-Low (HL) outliers, describing regions exhibiting high incidence rates that are bordered by regions of low incidence rates; and Low-High (LH) outliers, referring to regions with low incidence rates situated within contexts of neighboring high-incidence regions. Statistical significance for this analysis was determined at P < 0.05, supported by a 95% confidence interval.

Spatio-temporal scan analysis

To identify spatio-temporal clusters of influenza, the analysis in this study utilized monthly incidence rates as the unit for clustering, with counties serving as the finest spatial resolution. The spatial and temporal dimensions of cluster detection were constrained using a cylindrical scanning window, where the maximum allowable geographic extent was 25% of the study area, and the maximum time span was 50% of the total study period. A discrete Poisson model was used to estimate the projected case count for each scanning window based on the measured amount of cases and the population size inside versus outside the moving window (i.e., potential clusters) across candidate locations and time periods (14). The ratio of observed to expected case numbers inside versus outside the scanning window was used to calculate relative risk (RR). By analyzing observed and expected incidence rates, the log-likelihood ratio (LLR) was utilized to identify statistically significant clusters. The matching P value was obtained via a Monte Carlo simulation with 999 replications (19–21). The most likely cluster was determined to be the scanning window with the highest log-likelihood ratio (LLR) value. The Possion distribution’s LLR is calculated as follows:

Where n is the coefficient-adjusted predicted number of instances within the same window under the null assumption, C is the total case count, and c is the observed number of cases within the scanning window.

Spatio-temporal distribution characteristics of influenza incidence rates were generated by IBM SPSS 26.0 and Microsoft Excel 2021. Spatial auto-correlation analysis was performed by ArcGIS 10.8 (ESRI, Redlands, CA, USA). Spatio-temporal scan analysis was carried out with SaTScan 10.1.3 [22]. The cut-off point for statistical significance was established as a P value of less than 0.05. Supplementary Checklist 1 contains the completed STROBE Statement checklist for cross-sectional studies, detailing how each reporting item has been addressed in this study.

Results

Trends of the influenza incidence rates in Tianjin from 2010 to 2023

Between 2010 and 2023, the cumulative number of reported influenza cases in Tianjin reached 195,426. Figure 1 shows influenza incidence rates in Tianjin during the study period. Initially, the annual incidence rates were relatively stable, with minor fluctuations, from 2010 to 2019. A notable uptick occurred in 2020. This was followed by a sharp decrease in 2021. The most striking feature is the dramatic increase in 2023 (Fig. 1a). The monthly incidence rates for selected years, show a consistent pattern of low activity throughout most months, with a significant peak in February 2016 (Fig. 1b). Influenza incidence rates in all districts of Tianjin are consistent with an overall increase, which is particularly significant in 2019 and 2023 (Fig. 1c). The influenza cases in Tianjin showed an increasing trend year by year.

Fig. 1.

Influenza incidence rates in Tianjin (2010–2023). a: Annual influenza incidence rates in Tianjin from 2010 to 2023. b: Monthly influenza incidence rates in Tianjin by years. c: Intensity of influenza incidence rates across different districts in various years. The intensity scale is represented by a continuous color scale, from yellow (lowest intensity) to red (highest intensity)

Among these influenza cases, 99,546 cases are males and 95,880 cases are females. The sex ratio of males to females is 1.04: 1. Figure 2 shows the distribution of influenza cases by population and occupation in Tianjin from 2010 to 2023. In all age groups from 2010 to 2023, the cases were mainly 6–10 (43,907 cases, 22.5%) years old. The age groups of 0–5 and 6–10 years consistently represented a larger share of the total reported cases in each year (Fig. 2a). In terms of occupational distribution, students are the main group (73,132 cases, 37.4%), followed by preschool children (32,057 cases, 16.9%) and staff members (26,190 cases, 13.4%) (Fig. 2b).

Fig. 2.

Population and occupational distribution of influenza in Tianjin from 2010 to 2023. a: Percentage of influenza cases in different age groups from 2010 to 2022 in Tianjin. b: Occupational composition of the population affected by influenza

Spatial distribution of influenza incidence rates in Tianjin from 2010 to 2023

Influenza in Tianjin districts was generally on the rise, with active influenza incidence rates in 2019 and 2023 (Figs. 1c and 3). Figure 3 and Supplementary Fig. 1 present distinct spatial variations in influenza incidence rates across Tianjin, with notable shifts observed over the study period. Over a decade, the area with a high incidence rate was consistently concentrated in the eastern regions of Tianjin, and its level of concentration strengthened. Conversely, the incidence rate remained lower in Tianjin’s northern areas. Specifically, the eastern coastal regions of Tianjin (Binhai, Jinnan, Dongli, and Ninghe districts) have been the areas with high incidence rates for many years, and the inland areas in the northern regions of Tianjin continue to be at a low level.

Fig. 3.

Spatial distribution of influenza incidence rates in Tianjin. This shows the number of cases of influenza in each district in different years

Temporal distribution of influenza incidence rates in Tianjin from 2010 to 2023

Influenza in Tianjin generally exhibited seasonal characteristics. The influenza incidence rates were mainly concentrated from November to March of the following year (Fig. 1b). Binhai district has the highest frequency index of 0.85 and the most cumulative months of 143 (Supplementary Table 1). Nankai district has the highest duration index of 13.60, and Xiqing district has the highest intensity index of 3.91. The eastern region of Tianjin (Binhai, Jinnan, Dongli, and Ninghe districts) has all three prevalence indices at the top of the sixteen districts. The most epidemic waves are 20 months in Jinnan, while the fewest epidemic waves are 10 months in Nankai.

Spatial aggregation of influenza incidence rates in Tianjin from 2010 to 2023

Global spatial auto-correlation

In 2011 (Moran’s I = 0.14, Z = 2.10), 2016 (Moran’s I = 0.18, Z = 2.15), and 2018 (Moran’s I = 0.18, Z = 2.33), the respective Z-values above the crucial barrier of 1.96 for the distribution of normality at the threshold of significance of 0.05. These results clearly indicated the presence of spatial auto-correlation and clustering in influenza incidence rates during these years.

Local spatial auto-correlation

Between 2010 and 2023, Tianjin’s influenza incidence rates showed a distinct pattern in terms of their spatial distribution, with lower incidence rates in the northern regions and higher incidence rates along the eastern shore (Fig. 4 and Supplementary Fig. 2). Four different types of clustering were detected by the local spatial auto-correlation analysis for influenza incidence rates: high-high, low-low, high-low, and low-high clusters. The high-high clusters were mostly seen in the eastern part of Tianjin, namely in the districts of Jinnan and Binhai. Conversely, low-low cluster was concentrated in the northern region, particularly in Baodi district. Additionally, the eastern region’s vivinity is where low-high and high-low clusters are primarily found.

Fig. 4.

Local spatial autocorrelation of influenza incidence rates in Tianjin

Spatio-temporal scan analysis of influenza incidence rates in Tianjin from 2010 to 2023

Spatio-temporal scans were conducted to analyse influenza incidence rates in the two stages: from January 2010 to November 2019, and from December 2019 to December 2023. The results showed that influenza incidence rates exhibited significant spatio-temporal clustering. Table 1 presents the most likely cluster, as well as the secondary and tertiary clusters, identified during these analyses.

Table 1.

Spatio-temporal scan analysis of influenza incidence rates in Tianjin from 2010 to 2023

Study period	Cluster type*	Time frame	Radius (km)	Location included	RR	LLR	P-value
2010/01-2019/11	1	2017/12/1 to 2019/5/31	44.36	Binhai, Ninghe, Dongli, and Jinnan districts	7.73	6597.49	< 0.001
	2	2019/1/1 to 2019/4/30	13.73	Xiqing, Nankai, Hongqiao, and Beichen districts	11.05	2848.58	< 0.001
	3	2019/1/1 to 2019/4/30	3.89	Hedong, Hexi, and Heping districts	10.91	2571.67	< 0.001
2019/12-2023/12	1	2023/11/1 to 2023/12/31	43.92	Binhai, Ninghe, and Dongli districts	27.41	74011.43	< 0.001
	2	2023/11/1 to 2023/12/31	13.73	Xiqing, Nankai, Hongqiao, and Beichen districts	26.49	70005.70	< 0.001
	3	2023/11/1 to 2023/12/31	5.21	Hedong, Hexi, Heping, and Hebei districts	11.27	25889.35	< 0.001

*1: Most likely cluster. This refers to the most significant cluster identified in the analysis. It has the highest statistical significance, indicating that the recorded quantity of cases in this area and time period is significantly higher than expected, making it the most likely area to represent a true cluster. 2: Secondary cluster. These clusters have a lower significance compared to the most likely cluster but still hold statistical significance. 3: Tertiary cluster. These clusters have a lower significance than secondary clusters and may be potential areas of concern, but the statistical evidence is weaker

The spatio-temporal scan analysis’s findings, in terms of the time dimension, indicated that the influenza incidence rates in Tianjin from 2010 to 2023 are predominantly clustered during the winter and spring seasons. From the spatial dimension, the most likely cluster during the period from January 2010 to November 2019 was primarily distributed in the eastern region of Tianjin, encompassing four districts (Binhai, Ninghe, Dongli, and Jinnan districts). This cluster exhibited a RR of 7.73 and LLR of 6597.49 (P < 0.001). The temporal span of this cluster was from December 2017 to May 2019. Similarly, the most likely cluster during the period from December 2019 to December 2023 was also mainly distributed in the eastern region of Tianjin, covering three districts three districts (Binhai, Ninghe, and Dongli districts). This cluster demonstrated a significantly higher with a RR of 27.41 and LLR of 74011.43 (P < 0.001).

Discussion

There are a total of 150 countries and territories worldwide that are either island nations or possess coastlines, constituting more than three-quarters of the global country count (23). Coastal countries and cities often face complex public health challenges, particularly regarding prevention and management of infectious diseases, due to their unique geographical locations. Given the global prominence of coastal cities in the context of influenza epidemiology, the present study has comprehensively and systematically revealed the spatio-temporal epidemiology of influenza in Tianjin, a coastal mega-city of China, using surveillance data. This study, not only provided a method to delineate the spatio-temporal distribution of influenza but also identified high-risk areas, which is particularly pertinent for densely populated and economically crucial coastal regions. Our study’s temporal and spatial assessment of influenza was crucial in helping us understand the epidemic features of influenza in a coastal metropolitan environment. Furthermore, it has provided a scientific foundation for health policymakers, public health professionals, and clinicians to develop strategies for influenza prevention and control.

Analysis revealed an overall upward trend in influenza incidence in Tianjin between 2010 and 2023, aligning with the general pattern observed across China during the same period [24]. In contrast, influenza incidence rates decreased from a brief period of exceptionally high incidence rates in late 2019 to early 2023. From 2020 onward, the number of confirmed diagnoses decreased significantly compared to 2019, remaining in line with the United States, Singapore, and China [25–27]. Influenza rebounded after almost three years, peaking again in 2023. Trends in influenza incidence rates may be due to changes in a series of national policies. The People’s Republic of China’s National Health Commission published “Influenza Diagnosis and Treatment of Influenza (2019)” in 2019 [28], which helped standardize the procedures for diagnosing, treating, and reporting influenza, which in turn contributed to a rise in the reported case rate. In 2020, non-pharmaceutical interventions (NPIs) were enacted for the 2019 novel coronavirus disease (COVID-19) in China. These included measures like wearing masks, maintaining social distance, implementing travel restrictions, isolating patients, and similar public health interventions. By reducing contact with the influenza virus, they contributed to a decline in transmission [17, 21]. Since January 8, 2023, when the policy was introduced for the routine prevention and control of COVID-19 [29], the enhanced mobility of the population has provided a pathway for the spread of influenza. The influenza incidence rates in 2023 exceeded the highest record of the previous year in China.

In Tianjin, influenza exhibited an annual single-peak pattern, with the highest incidence occurring primarily in the first and fourth quarters. Population subgroups analysis identified students, office workers, and children in childcare settings as having elevated susceptibility. This aligns with prior studies [30–32] indicating that enclosed environments such as schools and workplaces facilitate influenza virus transmission and raise the likelihood of infection among individuals in these settings. After that, 33.56% of all instances included children under the age of 20. Therefore, it is recommended to strengthen ventilation and environmental cleaning in enclosed spaces like schools and companies. Then, students, staff, and childcare center children would be increased in influenza vaccination. Finally, it needs to encourage community support to reduce influenza incidence rates and protect high-risk groups.

To investigate how influenza incidence rates correlate geographically, influenza’s temporal and spatial spread in Tianjin between 2010 and 2023 was analyzed by combining spatial auto-correlation analysis and spatio-temporal scan analysis on the basis of drawing the map of influenza incidence rates in Tianjin. Spatial auto-correlation showed that the spatial evolution of influenza incidence rates were in the pattern of “high on the eastern coast”, and spatio-temporal scan analysis showed that the most likely cluster is was consistently concentrated in the eastern part of Tianjin. Two analyses consistently showed a high degree of clustering of influenza incidence rates in Tianjin’s eastern coastal areas. In Tianjin, influenza is less common in those 60 years of age or older and more common in teenagers between the ages of 0 and 10. In the eastern region of Tianjin (Binhai, Dongli, Jinnan, and Ninghe districts) (Supplementary Table 2), the occurrence of influenza in adolescents was more pronounced, accounting for about 50% of the overall count of cases. Over the years, the economic center of Tianjin has gradually shifted from the central region to the coastal economic region of the Binhai district. Geographically, such an industrial transition has created a transition zone between the old and new economic centers of Tianjin, namely Dongli, Jinnan, and Ninghe districts. The population of individuals aged over 65 has accounted for 67.5% of the total in the central region of Tianjin in 2022 [33]. More productive middle-aged people and adolescents are choosing to live and work in the transition zone. As a result, the reorganization and upgrading of the emerging industrial structure have brought extensive population movement and economic development to the eastern region of Tianjin. This also increased the risk of cross-regional transmission of influenza to a certain extent.

This study has several limitations. First, national influenza surveillance is based on passive reporting, which may result in underreporting, particularly among individuals who do not seek care at healthcare facilities. Second, the use of monthly surveillance data constrains the temporal resolution of our analysis. Although adequate for capturing seasonal trends, monthly indicators lack the granularity to reveal finer epidemic dynamics that weekly data could provide. Third, various factors, such as environment condition and surveillance methodologies, affect the influenza incidence rate; thus, future research should comprehensively examine how these variables affect observed incidence patterns. Finally, in the spatial auto-correlation analysis, we employed a binary contiguity-based spatial weight matrix, which does not account for continuous distance or population mobility patterns.

Future studies could extend this work in several directions. First, intergrating environmental covariates, such as temperature and humidity, through spatial regression or hierarchical Bayesian models would help disentangle the effects of climatic from those of surveillance practices. Second, replacing simple adjacency-based spatial weights with those derived from travel networks or human mobility data could more accurately capture transmission dynamics, particularly in urban settings. Third, systematically assessing the sensitivity of results to key parameters, such as the epidemic wave threshold or the chosen spatial scale, would enhance the robustness of the findings.

Conclusions

The influenza incidence rates in Tianjin from 2010 to 2023 displayed notable spatial and temporal aggregation patterns. A particular spatio-temporal cluster of influenza cases was identified in the eastern coastal region of Tianjin. The findings in the post-pandemic era revealed a significant increase in influenza cases, indicating a trend that warrants attention. It is crucial to ramain vigilant and enhance our efforts to prevent and control influenza outbreaks. The public health authoritie should implement comprehensive surveillance systems to closely monitor influenza activity. It is imperative that immunization programs be strengthened, especially for those at greatest risk including children and the elderly. Public health campaigns should educate the population on the importance of hygiene, vaccination, and early treatment seeking. Additionally, collaboration between health departments, businesses, and community leaders can facilitate a coordinated response to influenza outbreaks, ensuring that control measures are implemented promptly and effectively. The study also emphasizes the need for a proactive approach in regions with rapid economic development, high population density, and significant foreign trade and population mobility.

In summary, this study provides a theoretical framework for influenza prevention and control in coastal mega-cities. As these regions undergo economic development and industrial restructuring, it is important for local health departments to focus on the transmission of influenza not only in coastal economic areas but also in the transitional zones between older and newer economic centers.

Authors’ contributions

JL and ZM jointly conceived the study, collected and analyzed the data, and drafted the initial manuscript. LL and XD participated in data curation and analysis. YL and JY assisted with data collection and organization. ZL and XS contributed to reviewing and revising the manuscript. XY and YZ were responsible for the conceptual design and overall supervision of the study. CC secured funding, conceived and coordinated the study, and provided supervision and guidance throughout the research process and manuscript preparation. All authors read and approved the final manuscript.

Funding

This study was supported by the Training program of public health talents in China.

Data availability

The data presented in this study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Tianjin University (Approval No. TJUE-2024-004). The requirement for obtaining informed consent was waived by the committee as the research involved a retrospective analysis of fully anonymized public health surveillance data, in accordance with applicable guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.