Spatial, temporal, and spatiotemporal analysis of pertussis in Shapingba District, China, 2014–2024

Abstract

Background

Pertussis has resurged globally, with a sharp increase in 2024 posing serious risks to children health. This study analyzed the epidemiological and spatiotemporal patterns of pertussis in Shapingba District, China, from January 1, 2014 to December 31, 2024, to identify high-risk groups and areas for targeted prevention.

Methods

Reported pertussis cases were obtained from China Information System for Disease Control and Prevention. Descriptive statistics, spatial autocorrelation (GeoDa), and spatiotemporal scan analysis (SaTScan) were used. ArcMap was applied for visualization.

Results

A total of 2,899 cases were reported, with an average annual incidence of 18.42 per 100,000, peaking at 157.26 in 2024. An age shift was observed: before 2021, infants under 1 year had the highest incidence; from 2021, cases rose among children aged 6–13 years. Spatial clusters were identified in 2015, 2016, 2022, and 2024, with high-risk areas shifting west and southeast over time. Eleven significant clusters occurred between March and August 2024.

Conclusion

Pertussis in Shapingba District showed increasing incidence and shifting age and spatial patterns. Enhanced surveillance, optimized immunization schedules, and targeted interventions are urgently needed.

Introduction

Pertussis, caused by Bordetella pertussis, has affected humans for at least 800 years. It is a highly contagious disease that can be fatal, particularly in unvaccinated individuals and young infants [1]. Prior to the introduction of vaccines, pertussis posed a significant global health burden, with over 300 million children estimated to have died from the disease annually. Although the global morbidity and mortality associated with pertussis declined significantly following the implementation of the World Health Organization’s Expanded Program on Immunization (EPI), a resurgence of the disease has recently been reported in several countries, including the United States, Canada, and Australia, despite high vaccination coverage [2–4].

China once experienced a high burden of pertussis, with reported incidence rates exceeding 100 cases per 100,000 population prior to the introduction of vaccination [5]. Since 1978, the nationwide implementation of pertussis vaccination has achieved and sustained high immunization coverage. As a result, reported cases of pertussis declined substantially, with the incidence dropping to below 1 per 100,000 by the late 1990s. Between 2006 and 2013, fewer than 3,000 cases were reported annually across the country, with fewer than 10 associated deaths [6]. However, since 2013, China has experienced a continuous rebound in incidence, culminating in an explosive surge in 2024, with approximately 0.49 million pertussis cases reported, a 12.3-fold increase compared to the previous year (https://www.chinacdc.cn/). Consistent with the China trend, Chongqing Municipality also showed a sharp upward trajectory in pertussis incidence, surging from 0.25 per 100,000 population in 2014 to 70.82 per 100,000 in 2024. This level ranked among the top three provinces or municipalities nationwide and was significantly higher than the national average [7]. Recent studies further indicate significant geographical heterogeneity among Chongqing’s 38 districts and counties, with Shapingba District consistently reporting incidence rates above the municipal average and ranking second in relative risk [7]. Furthermore, recent studies have revealed new epidemiological patterns in the distribution of pertussis cases, with the proportion of cases occurring in adolescents and adults increasing annually [8, 9]. These emerging epidemiological characteristics and significant regional differences indicate that a refined, regional analysis of pertussis incidence is necessary.

Advanced spatiotemporal epidemiological research methods are crucial for effectively responding to emerging and re-emerging infectious diseases, as these methods can identify high-risk areas and time periods of disease transmission [10]. This analytical paradigm has been widely applied in the surveillance and control of major infectious diseases. For example, the studies by Rancati et al. and Podrecca et al. employed multi-regional time-series models to uncover bidirectional and lagged associations between COVID-19 policies, vaccination, and key epidemiological indicators, thereby capturing dynamic changes across multiple spatial units and providing quantitative evidence to support public health decision-making [11, 12]. Concurrently, spatiotemporal clustering methods have become essential tools for analyzing the prevalence of various infectious diseases (such as dengue and chikungunya), offering targeted guidance for disease control and prevention [13, 14]. This correlation-based spatiotemporal analytical approach is particularly critical for a disease like pertussis, which is characterized by significant geographical heterogeneity and a resurgence trend. However, previous pertussis research in China has primarily focused on descriptive epidemiological features, with relatively few studies addressing the regional spatiotemporal distribution, making it difficult to reveal the dynamic changes of the disease in the geographical and temporal dimensions [15].

Therefore, to address this gap in the knowledge of pertussis spatiotemporal patterns in Shapingba District, this study employs spatial autocorrelation and spatiotemporal scan statistics to evaluate the statistical significance of clustering from January 1, 2014, to December 31, 2024 and to delineate the extent of epidemic hot spots. The findings will contribute to a comprehensive understanding of the epidemiological characteristics and spatiotemporal distribution patterns in the Shapingba District, providing a scientific basis for formulating targeted health policies and rationally allocating medical resources.

Methods

Study area

Chongqing is a municipality located in southwestern China. Shapingba District, situated in the western part of Chongqing’s central urban area, comprises 26 towns and subdistricts. It serves as the educational and cultural hub of the municipality, with a population of approximately 1.49 million (Fig. 1).

Fig. 1.

The geographic location of Shapingba District, Chongqing, China. (The map was created with ArcGIS software version 10.8.1)

Data collection and management

Data on pertussis cases reported in Shapingba District between January 1, 2014, and December 31, 2024, were obtained from the China Information System for Disease Control and Prevention (CISDCP). To ensure a consistent case definition, the diagnostic criteria in the Diagnosis and treatment plan for pertussis (2023 Edition) were applied uniformly throughout the study period [16]. Case classification was based on epidemiological history, clinical manifestations, and laboratory testing, comprising both laboratory-confirmed and clinically diagnosed cases.

Reported case information included a unique case identification number, administrative code, date of symptom onset, and demographic characteristics, such as sex, age, occupation, and residential address. Case report forms were completed by qualified physicians and submitted to the CISDCP within 24 h by trained personnel, in accordance with the Law of the People’s Republic of China on the Prevention and Treatment of Infectious Diseases. Population data were obtained from the Shapingba District Statistical Yearbook and the Shapingba District Center for Disease Control and Prevention.

Spatial autocorrelation analysis

GeoDa software 1.18 was used to conduct global spatial autocorrelation (GSA) and local spatial autocorrelation (LISA) analyses. GSA characterizes the overall spatial pattern of the region, reflecting the degree of association between attribute values and their spatial locations [17]. Global autocorrelation was assessed using Moran’s I, which ranges from − 1 to 1. A Moran’s I > 0 indicates a positive spatial correlation, Moran’s I = 0 suggests a random spatial distribution, and a value < 0 indicates a negative spatial correlation. The closer the index is to 1 or − 1, the stronger the spatial autocorrelation and the more evident the spatial clustering [18]. The Z value is used to test whether there is spatial autocorrelation between regions, with α = 0.05 as the test level. A p < 0.05 indicates statistically significant global spatial autocorrelation [18]. LISA was employed to assess the spatial correlation between pertussis incidence in each subregion and its neighboring areas. Based on LISA results. LISA is used to divide the clusters into four cluster association patterns: “high-high”, “high-low”, “low-high”, and “low-low“ [19]. The “high-high” and “low-low” clusters (positive LISA values) are interpreted as spatial clusters, while the “high-low” and “low-high” clusters (negative LISA values) are considered spatial outliers.

Spatial and space-time cluster analysis

Spatial autocorrelation analysis can identify disease spatial heterogeneity, a limited function for detecting the specific extent of disease clustering and its temporary clustering. Therefore, this study conducted a spatiotemporal cluster analysis. The analysis was performed using SaTScan 10.1 software, with street serving as the minimum spatial units. Monthly pertussis data were used to perform spatiotemporal clustering analysis. SaTScan accounts for both spatial and temporal dimensions by employing a cylindrical scanning window that dynamically adjusts across space and time [20]. In this study, a circular window shape was selected, which is widely used in spatiotemporal epidemiological research and appropriate for the relatively small geographic scale of Shapingba District [21]. The retrospective space–time scan statistic was applied using a Poisson probability model with a discrete likelihood ratio test. Based on prior studies, the maximum spatial window was set to 10%-30% of the total population, and the maximum temporal window was set to 50% of the study period [15]. To assess statistical significance, a Monte Carlo simulation with 9,999 replications was performed. Spatiotemporal clustering was considered significant if the maximum log likelihood ratio (LLR) was calculated with P < 0.05. The higher the LLR and relative risk (RR) values, the stronger the degree of clustering. The scanning window with the highest LLR value was designated as Type I clustering, while other statistically significant windows were classified as Type II clustering areas [22]. The significance level was set at α = 0.05.

Statistical software

GSA and LISA analyses were conducted using GeoDa software 1.18. Spatiotemporal scanning was performed using SaTScan software 10.1. All spatial analysis results were visualized using ArcGIS software (version 10.8.1, ESRI, Inc., Redlands, CA, USA).

Ethics statement

Pertussis data were extracted from annual reports by the Shapingba District Center for Disease Control and Prevention, which provided aggregated counts of patients reported by street and month. All data were anonymized.

Results

General demographic characteristics

A total of 2,899 pertussis cases reported between January 1, 2014 to December 31, 2024 were included in this study (Fig. 2). The average annual incidence of pertussis has been increase since 2014, peaking in 2019, with an incidence rate of 0.99 per 100,000-10.16 per 100,000 from 2014 to 2019. After 2019, the incidence of pertussis declined and then rebounded from 2022 onwards. This was followed by a sharp increase in incidence in 2024, to 157.26 per 100,000. The temporal distribution of cases exhibited clear seasonal patterns, with a pronounced peak from April to August. In terms of population characteristics, the overall gender ratio (male: female) was approximately 1:1.01. The age distribution changed remarkably, in 2014–2021, the highest number of patients were < 1 year old (infancy) (60.00% to 90.00%), followed by patients 1–3 years old, about 18.63%; patients 3–6 years old, about 7.60%, patients 6–13 years old, about 4.58%, and patients 13–18 years old, about 0.85%. However, from 2021 to 2024, the number of patients aged 6–13 years increased significantly, with the highest percentage of 60.80%, followed by patients aged 1–3 years, about 20.21%, and infants < 1 year about 7.86%, patients 3–6 years old about 4.77%, patients 13–18 years old about 2.76%. Similarly, the distribution of cases across different population groups changed considerably. 2014–2021, the patients were mainly children living in the diaspora, accounting for about 90.03%; nursery children and students accounted for about 7.34% and 2.10%, respectively. In contrast, in 2022–2024, the number of cases of students and nursery children increased, accounting for about 47.93% and 33.96%, respectively; scattered children accounted for about 14.50%.

Fig. 2.

Epidemiological characteristics of pertussis in Shapingba District from 2014 to 2024. a Monthly distribution of pertussis. b Annual incidence of pertussis in different gender. c Occupation distribution of pertussis. d Age distribution of pertussis

Geographical distribution

The geographic distribution of pertussis incidence rates in Shapingba District from January 1, 2014 to December 31, 2024 is shown in Fig. 3. The average annual incidence rates across the district remained low (0.99–5.00 per 100,000) during 2014–2017, 2020–2021 and 2023. However, regional increases were observed in 2018, 2019, 2022 and 2024, particularly in the western areas of Shapingba District, including Xiyong, Chenjiaqiao, Xianglushan, Huxi, and Tuzhu, as well as in the southeastern areas such as Qinjiagang, Shijingpo, and Shuangbei, which recorded higher average annual incidence rates compared to other regions. Notably, in 2024, the incidence rate of pertussis increased dramatically across all subdistricts of Shapingba District, with the western and southeastern regions continuing to be the primary high-prevalence areas.

Fig. 3.

Annual incidence of pertussis by street in Shapingba District from 2014–2024

Spatial autocorrelation analysis

The results of the spatial autocorrelation analysis of the average annual incidence of pertussis in Shapingba District are presented in Table 1; Fig. 4. Statistically significant spatial clustering was observed in the years 2015, 2016, 2022, and 2024 (P < 0.05). The LISA analysis effectively identified high-high (hot spots) and low-low (cold spots) clusters of pertussis incidence. It can be seen from the LISA graph that the location of the hot spots of pertussis had a tendency to shift, and the hot spots in 2015 the high-high cluster (dark red) was located in the Qinjiagang, the high-high cluster area in 2016 was in the north Huilongba and Zhongliang, and the high-high cluster areas in the other 2 years were located in the western region, including Xianglushan, Zengjia and Chenjiaqiao.

Table 1.

The Spatial autocorrelation analysis of pertussis in Shapingba District, Chongqing, China, 2014–2024

Year	Moran I	E[I]	SD	Z-value	P-value
2014	0.0393	-0.0417	0.1426	0.5468	0.29
2015	0.3535	-0.0417	0.1334	2.9822	0.02
2016	0.3227	-0.0417	0.1315	2.9481	0.01
2017	-0.0706	-0.0417	0.1349	-0.2314	0.44
2018	-0.0969	-0.0417	0.1401	-0.4625	0.33
2019	-0.1120	-0.0417	0.1326	-1.0207	0.30
2020	-0.1825	-0.0417	0.1378	-1.10156	0.16
2021	0.0852	-0.0417	0.1290	0.9664	0.17
2022	0.296	-0.0417	0.1377	1.9996	0.03
2023	0.0148	-0.0417	0.1283	0.4178	0.33
2024	0.1695	-0.0417	0.1090	1.9710	0.04

Fig. 4.

Yearly Local Indicators of Spatial Association (LISA) cluster maps for pertussis incidence, Shapingba District, 2014–2024

Spatiotemporal clusters analysis

The results of the spatiotemporal cluster analysis of pertussis incidence are summarized in Table 2. A total of 11 statistically significant clusters are identified. The primary cluster was located in the western part of Shapingba District, encompassing the subdistricts of Chenjiaqiao, Huxi, and Xianglushan, with a time window from March 1, 2024, to August 31, 2024. The secondary clusters1 was detected in the eastern region, including Shapingba, Ciqikou, Yubeilu, Tuwan, and Xiaolongkan, spanning the period from April 1, 2024, to July 31, 2024. A third cluster was identified in the eastern subdistricts of Xinqiao, Qinjiagang, and Tianxingqiao, with a time range from March 1, 2024, to July 31, 2024.

Table 2.

The clusters of pertussis cases detected by using the purely spatial scan statistics

Cluster type	Date(Year/Month/Day)	Area	Observed	Expected	RR	LLR	P value
Most likely	2024.3.1–2024.8.31	Chenjiaqiao, Huxi, Xianglushan	646	28.67	22.53	1466.61	< 0.01
Secondary	2024.4.1–2024.7.31	Shapingba, Ciqikou, Yubeilu, Tuwan, Xiaolongkan	456	19.1	23.87	1044.8	< 0.01
2nd Secondar	2024.3.1–2024.7.31	Xinqiao, Qinjiagang, Tianxingqiao	317	21.8	14.54	569.02	< 0.01

Discussion

This study used GIS and scan statistics to systematically reveal the epidemiological characteristics and the spatial, temporal, and spatiotemporal clustering patterns of pertussis in Shapingba District from January 1, 2014, to December 31, 2024. The findings indicate that the annual incidence rate and reported cases of pertussis showed an overall upward trend, with a significant surge observed in 2024. The incidence dramatically increased from 0.99 per 100,000 in 2014 to 157.26 per 100,000 in 2024, with 2,340 cases reported in that single year, exceeding the cumulative cases of the preceding decade (559 cases). The peak incidence period for pertussis occurs between April to August. Furthermore, although infants and young children remain a high-risk group for pertussis infection, the proportion of cases among older children has significantly increased since 2022.

The sharp increase in pertussis incidence in Shapingba District is highly consistent with the national trend observed in China. According to surveillance data from the Chinese Center for Disease Control and Prevention, approximately 0.49 million pertussis cases were reported in 2024, 12.3 times higher than in 2023 (https://www.chinacdc.cn/). Moreover, several high-income countries, including the United States, Japan, South Korea, Australia, and member states of the European Union, also experienced substantial increases in pertussis incidence in 2024, ranging from 4.0-fold to 161.3-fold compared with 2023 (http://www.chinawatch.cn/), showing the breadth and severity of the outbreak. Several factors likely contributed to the sharp rise in Shapingba District. First, during the COVID-19 (2020–2023), stringent non-pharmaceutical interventions (NPIs), such as mask-wearing and social distancing, not only suppressed pertussis transmission but also limited opportunities for natural boosting of immunity, leading to an accumulation of susceptible individuals. Given that the basic reproduction number (R₀) of pertussis is estimated at 12–17, the pathogen has substantial transmission potential [3]. Consequently, after these control measures were relaxed in 2024, increased population mobility, school reopening, and the resumption of social activities created favorable conditions for rapid spread. Second, the COVID-19 accelerated the adoption of molecular diagnostic techniques. Since 2020, medical institutions in Shapingba District routinely implemented polymerase chain reaction (PCR) testing for patients presenting with coughs, substantially improving case detection and suggesting that the true burden of pertussis may have been underestimated in previous years. Third, following the lifting of control measures in 2024, most residents in the district experienced COVID-19 infection. Evidence suggests that COVID-19 may increase susceptibility to pertussis by disrupting respiratory epithelial barriers or inducing immune suppression [23]. In summary, weakened immunity and the accumulation of susceptible populations, the resumption of social activities, enhanced diagnostic capacity, and COVID-19 related increases in susceptibility likely acted synergistically to drive the dramatic resurgence of pertussis in Shapingba District in 2024.

This study found an upward shift in the age of pertussis onset in Shapingba District, and the proportion of cases among nursery children and students continued to increase. Similar trends have been observed in in other parts of China. Cai et al. reported that in 2024, children aged 4–10 years accounted for 44.0%–78.5% of all pertussis cases across five provinces in China [24]. Several factors may explain this shift. First, changes in vaccine strategy and waning immunity are critical contributors. Since 2008, China has replaced the whole-cell pertussis vaccine (wP) with the acellular pertussis vaccine (aP). Evidence suggests that the effectiveness of aP declines to 50%–70% within 4–7 years after completion of the primary series [25]. By 2021, many children in our study population had reached the period when protection from their last scheduled dose had begun to decline, leaving them increasingly vulnerable. Second, school-based outbreaks have become a growing concern. In 2024, Shapingba District recorded approximately 30 school-associated pertussis outbreaks, far exceeding the 1–3 outbreaks typically reported in previous years, which likely amplified transmission among children and contributed substantially to the sharp increase in this age group. Third, natural infection does not confer lifelong immunity, leaving a persistent risk of reinfection. Finally, pertussis in older children often presents with atypical or mild symptoms and was easily overlooked in the past. The increased diagnostic sensitivity, partly due to the widespread adoption of PCR, has markedly improved case recognition in this age group, which also contributes to the increased proportion of older cases. Although the proportion of infants among overall cases has declined in recent years, their immature immune systems render them highly vulnerable to severe disease and complications, highlighting the continued need for focused protection in this group. To address these challenges and reinforce herd immunity, China revised its immunization schedule in January 2025, introducing a three-dose primary series at 2, 4, and 6 months, followed by boosters at 18 months and 5 years. Our research team will continue to monitor the epidemiological trends of pertussis following the implementation of this new strategy to assess its impact on disease burden.

Spatial analysis revealed that the incidence of pertussis in Shapingba District exhibited non-random spatial distribution, with evident spatial clustering. LISA analysis indicated significant spatial heterogeneity, with high-high clusters primarily concentrated in the western part of Shapingba District, including Zengjia, Chenjiaqiao, and Xianglushan streets, indicated that these three administrative units were hotspots for the regional spread of pertussis. Several factors may have contributed to this phenomenon. First, as the development centers of Shapingba District, these streets have attracted large numbers of migrant workers due to industrial expansion, resulting in a large and highly mobile population that is prone to community transmission. High population mobility may also lead to incomplete or untimely vaccination. Second, as the development of the focus area, more new primary schools and child care institutions, leading to a dense concentration of school-age children, the confined places to accelerate the droplet transmission of Bordetella pertussis. Additionally, the concentration of colleges and universities in the streets of Xianglushan and Chenjiaqiao means children of faculty and staff are more likely to be diagnosed through proactive screening, primarily due to higher parental health literacy and easier access to healthcare services. These factors may collectively contribute to the observed clustering patterns of pertussis within the district.

Spatio-temporal clustering analysis identified three primary clusters located in the western and southeastern regions of Shapingba District, with clustering periods occurring between March and August 2024. The annual incidence trends and seasonal changes showed that the incidence of pertussis in Shapingba District in 2024 was much higher than in other periods, suggesting that the situation of pertussis prevention and control was critical. Therefore, local public health or medical institutions should develop and implement targeted intervention strategies before the incidence peak according to the seasonal characteristics of pertussis.

In conclusion, targeted prevention and control measures should be implemented based on the epidemiological and spatiotemporal patterns identified in this study to further reduce pertussis transmission. First, community-level immunization should be reinforced, ensuring timely administration of DTaP booster doses to eligible children to counter waning immunity and the accumulation of susceptible individuals. Second, the observed seasonal clustering of cases between April and August highlights the need to strengthen surveillance and implement preventive interventions ahead of the epidemic peak, including school-based health education, enhanced respiratory hygiene, and early identification of suspected cases in kindergartens and primary schools. Third, the spatial aggregation of pertussis in western Shapingba District underscores the importance of prioritizing high-risk areas, particularly communities with dense migrant populations and educational institutions, where vaccination and catch-up immunization should be intensified, mobile children closely monitored, and absenteeism due to respiratory illness tracked. Finally, closer collaboration between educational and public health authorities is recommended, focusing on improved ventilation, strengthened environmental hygiene, and targeted training for teachers on common infectious diseases, thereby reducing the risk of pertussis and other respiratory infections.

However, this study also has some limitations. First, the data were obtained from China’s passive surveillance system, which means that some pertussis cases may have been unreported because of their subclinical symptoms, although quality control of data collection has been an important part of China’s disease surveillance system for all reported diseases. Second, as a retrospective study, it was difficult to obtain environmental, social, and other potential determinants at the subdistrict level. Therefore, our analysis was limited to describing the spatial distribution patterns and epidemiological characteristics of pertussis from January 1, 2014 to December 31, 2024, and the interpretation of causal relationships remains insufficient. Third, regarding the sharp increase in 2024, we were unable to obtain serological data or strain characterization information, which limited our ability to directly assess waning immunity or pathogen variation. Therefore, while several plausible explanations have been proposed, the precise underlying mechanisms could not be fully elucidated.

Conclusion

To the best of our knowledge, this study is the first to characterize whooping cough epidemics in Shapingba District from January 1, 2014 to December 31, 2024 using spatio-temporal analysis. The distribution of pertussis epidemics in Shapingba District has a certain spatial and temporal aggregation, mainly in the western and southeastern regions, and the aggregation time is from April to August. This study also found that the incidence of pertussis in older children has increased significantly in recent years. Therefore, according to the changing characteristics and epidemiological trends of pertussis, the arrangement of health resources for the prevention of pertussis should be adjusted, the priority prevention areas should be divided appropriately, and focused preventive measures should be implemented.

Abbreviations

EPI

Expanded Program on Immunization.

GSA

Global spatial autocorrelation

LISA

Local spatial autocorrelation

CISDCP

China Information System for Disease Control and Prevention

Author contributions

JCS participated in the study design and revised the manuscript. QZ participated in the study design and revised the manuscript. SRF carried out data collection, data analysis and drafted the manuscript. BY participated in the revision of the article. All authors read and approved the final manuscript.

Funding

The study was supported by 2025 Chongqing public health key specialty (discipline) project and the Chongqing Shapingba District Technology Innovation Project (grant number 2025188).

Data availability

The datasets generated during and analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study involving human participants was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was reviewed and approved by the Ethics Committee of the Shapingba District Center for Disease Control and Prevention. As this study was based on routine surveillance data and involved no direct contact with participants, informed consent was waived by the Ethics Committee.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.