Pertussis, an acute respiratory infection caused by Bordetella pertussis (B. pertussis) (1), has afflicted humans for centuries (2). This highly contagious disease impacts individuals across all age groups and can be fatal in unvaccinated populations, particularly in young infants (3).

Recent reports indicate a “resurgence” of pertussis (4–6), accompanied by substantial outbreaks globally (6–8). Pertussis has reemerged as a major public health threat, particularly for infants and young children (3, 9). Despite sustained high coverage (>99%) of the three-dose diphtheria-tetanus-acellular pertussis (DTaP) vaccine series in China over the past decade, the pertussis notification rate has exhibited a fluctuating but overall increasing trend (6). Pertussis remains a persistent public health challenge (6). China is one of the countries with the highest incidence of pertussis in recent years (1, 6).

Chongqing, a municipality in southwestern China, has experienced a dramatic 14.59-fold increase in its pertussis notification rate, from 0.25 per 100,000 in 2004 to 3.66 per 100,000 in 2018 (10). This rate exceeds the national average and imposes a substantial medical economic burden (10, 11).

Understanding pertussis transmission patterns, particularly spatial and temporal dynamics, is crucial for effective control. However, such analyses have been conducted in various regions (12–16) and in other Chinese provinces (17, 18). These patterns are highly region-specific, influenced by local demographic, environmental, and surveillance factors. Therefore, despite the valuable insights from previous studies, there remains a critical need for localized analyses to inform targeted public health interventions in specific high-burden areas. This study aimed to investigate the shifting epidemiology of pertussis in Chongqing from 2013 to 2024 by identifying priority populations, analyzing long-term trends, and detecting high-risk spatiotemporal clusters, thereby providing an evidence-based foundation for optimizing local control strategies.

Our 12-year longitudinal analysis reveals a critical epidemiological shift in pertussis transmission dynamics in Chongqing. Over the 12-year study period (2013–2024), infants (aged <1 year) consistently had the highest notification rates, which surged from 17.25 to 1027.14 per 100,000 person-years. Notably, infants aged <3 months accounted for 25% of all reported cases during the 2013–2023 study period, a trend consistent with surveillance data from multiple provinces across China (17, 18).

This vulnerability arose from China’s pre-2025 immunization schedule, under which infants aged <3 months had no vaccine-induced immunity, while those aged 3–12 months had incomplete protection during the primary series (38). Furthermore, the underrecognition of adult pertussis due to widespread self-medication and low clinical suspicion creates a hidden reservoir of infection, elevating the risk of household transmission to young infants (39). As a result of this heightened exposure risk and their own susceptibility, the majority of severe cases occur in infants aged < 3 months, as emphasized by Kandeil et al. (3), highlighting a critical window of susceptibility before vaccine eligibility. Maternal Tdap vaccination during pregnancy is therefore a vital strategy for bridging this immunity gap (40). A US case–control study demonstrated that prenatal Tdap vaccination administered between 27 and 36 weeks of gestation was 92.5% effective (95% CI: 38.5–99.1) in preventing pertussis in infants <2 months (41). Recognizing its value, over 40 countries have now implemented recommendations for maternal Tdap immunization, albeit with variations in the recommended gestational window (3). In response, China has adjusted its national immunization schedule, effective in 2025, which advances the first DTaP dose from 3 to 2 months of age and introduces a booster dose at 6 months old (22). While China has not yet introduced maternal Tdap immunization, this schedule revision represents a significant step towards reducing the vulnerability of young infants.

Our data strongly suggest that school-aged children (5–10 years) became the core transmitters driving community spread. This is supported by their pivotal role in initiating intense spring–summer epidemics from 2021 onward, culminating in 2024 when they constituted 60.89% of all reported cases—7.4 times the case count in infants (42). The subsequent emergence of distinct seasonal peaks in adolescents and adults after 2022 demonstrates the expansion of transmission from this school-aged reservoir, marking the establishment of sustained community-wide transmission chains (43). Our analysis revealed distinct age-dependent sex differences in pertussis incidence. A significant male bias was observed in infants under 1 year (IRR = 1.11, 95% CI: 1.06–1.16), which contrasts with the female bias found in adults (IRR = 0.64, 95% CI: 0.57–0.71) and preschool-aged children (3–5 years; IRR = 0.88, 95% CI: 0.82–0.95). This pattern of highest male susceptibility in infancy aligns with global epidemiological data showing the greatest heterogeneity in sex-specific incidence during the first year of life (44). The biological plausibility of this finding is supported by evidence that male infants have a well-documented increased susceptibility to respiratory infections, likely due to sex-based differences in early immune maturation (45).

The substantial increase in reported pertussis cases in 2024, particularly among school-aged children, warrants careful interpretation. Our ITSA identified a significant level change coinciding with the 2024 diagnostic criteria revision. The observed pattern, with school-aged children (5–10 years) constituting the majority of cases, while infants exhibit the highest notification rate, suggests that school-aged children now represent a key reservoir for B. pertussis transmission in Chongqing. However, it is critical to interpret this surge as resulting from both substantially enhanced surveillance sensitivity and a genuine underlying increase in pertussis transmission. This genuine increase may have been exacerbated by an accumulation of susceptible individuals during the COVID-19 pandemic, as evidenced by the pre-existing upward trends in these age groups.

Three strands of evidence converge to support this interpretation of surveillance enhancement superimposed on a genuine transmission increase. Primarily, the immediate and substantial surge in notifications, particularly for laboratory-confirmed cases, corresponds precisely with the implementation of the revised criteria. Furthermore, the epidemiological dynamics before 2024, characterized by an initial non-significant increase (APC = 113.41%, 2013–2017) followed by a period of non-significant decrease (APC = -14.15%, 2017–2023) in pertussis notification rate, indicate that underlying transmission was active and fluctuating, rather than following a stable or consistently declining trajectory. Finally, the characteristic shift in the age distribution in 2024 aligns with the specific enhancements to the diagnostic criteria aimed at improving detection in older age groups. While we cannot definitively quantify the relative contributions of improved detection versus factors such as the accumulation of susceptibles, the evidence strongly suggests that the diagnostic revision revealed a substantial and growing disease burden that was previously under-ascertained.

The pronounced increase in laboratory-confirmed cases is consistent with established knowledge that enhancements in diagnostic sensitivity can lead to improved case detection (46). This context is particularly relevant given that the infrastructure for molecular diagnostics, widely established for SARS-CoV-2 testing, is readily adaptable for the detection of other respiratory pathogens like B. pertussis (47). Furthermore, the sustained upward trajectory in cases aligns with the well-documented global resurgence of pertussis, which is largely attributed to the waning of immunity following pertussis vaccination (48, 49). This pattern of resurgence is not unique to Chongqing but reflects broader epidemiological trends reported in national surveillance data from China (50, 51) and aligns with global trends of the resurgence of pertussis, driven by age-group shifts (6). This epidemiological transition is likely driven by waning immunity following childhood vaccination. A recent study attributed China’s resurgence of pertussis largely to this demographic transition, characterized by a rising proportion of cases among older children (52).

Beyond the shifting age distribution, we observed a significant overall resurgence of pertussis in Chongqing from 2013 to 2024, with notification rates increasing from 0.20 to 71.38 per 100,000 person-years (AAPC = 32.16%). The parallel increase in laboratory-confirmed cases (AAPC = 30.31%) indicates an underlying rise in disease transmission, while the dramatic surge in 2024 likely represents the combined effect of this genuine resurgence and substantially enhanced surveillance sensitivity through the diagnostic criteria revision. This upward trend reflects the widely recognized resurgence of pertussis throughout China (6, 53).

Multiple interconnected mechanisms likely contribute to the observed resurgence of pertussis in Chongqing. Foremost is the waning of immunity following both vaccination and natural infection (2, 51, 54). Protection from aPVs wanes at an estimated rate of 9.6% (95% CI: 8.6–10.6) per year after completion of the primary series (55), substantially expanding the susceptible population over time (51, 54, 56). A systematic review confirmed this progressive waning, with the relative risk of infection increasing from 0.15 (95% CI: 0.09–0.26) at <1 year post-vaccination to 0.41 (95% CI: 0.30–0.56) at 5–7 years (57). Additionally, although the transition from whole-cell (DTwP) to DTaP vaccines reduced adverse effects, it may have contributed to resurgence due to less durable immunity (5). Furthermore, genomic evolution of B. pertussis under acellular vaccine pressure may also facilitate pathogen re-emergence. This evolution is characterized by emerging allelic variations that enhance immune evasion (58).

The patterns we observed in our age-stratified analysis are consistent with these resurgence mechanisms. Our analysis identified the most rapid increases in notification rates in school-aged children (5–10 years: AAPC = 94.91%) and adolescents (10–18 years: AAPC = 89.32%). Similar trends have also been observed in previous studies (57, 59). While we did not find direct evidence supporting waning immunity as a driver within our model, the observed trends are compatible with this established mechanism. However, the extraordinary magnitude of the surge in 2024 coincides temporally with the implementation of revised diagnostic criteria, suggesting that enhanced case detection substantially contributed to the observed increase. The legacy effect of historical vaccination practices, as posited by Domenech de Cellès et al. (60), whereby widespread whole-cell vaccination reduced natural immune boosting, may serve as an additional contributing factor. In China, the DTaP antigen combination forms the basis of the national immunization program. It is administered both as a standalone DTaP vaccine and as a component in multivalent vaccines, which may also include inactivated poliovirus (IPV) and Haemophilus influenzae type b (Hib) antigens (61). Compared to the pre-2025 schedule, China’s 2025 revision advances the first DTaP dose from 3 to 2 months, delays the third primary dose from 5 to 6 months, extends the interdose interval from one to two months, and replaces the DT booster at 6 years of age with a DTaP vaccine (22). Our interrupted time-series analysis reveals the profound impact of revisions to China’s national diagnostic criteria on pertussis surveillance data in Chongqing. The attenuated notification rate following the 2017 guideline implementation (25), which mandated laboratory confirmation through serological testing and PCR, likely reflects improved diagnostic specificity that reduced false-positive reporting. Most strikingly, the immediate 62.1-fold surge in laboratory-confirmed cases following the 2024 revision corresponds directly to the 2023 national protocol (26) that introduced clinical epidemiological criteria and enhanced molecular detection capacity. This magnitude of increase underscores the profound impact of surveillance changes, though it likely captures a combination of previously undetected cases and genuine epidemiological growth.

These divergent trends demonstrate that both surveillance changes and genuine epidemiological shifts can profoundly influence observed pertussis patterns, highlighting the importance of considering multiple factors when interpreting surveillance data. This underscores the critical importance of accounting for changes in diagnostic policy when interpreting pertussis trends over time in China (25, 26).

Our multilevel model identified higher district-level income as an independent predictor of elevated pertussis notification rates (β = 0.56, p < 0.001), consistent with literature linking socioeconomic status (SES) to disease surveillance sensitivity (2). Specifically, higher-SES populations have greater access to primary care and fewer financial barriers to seeking care when symptomatic, factors that enhance the likelihood of pertussis diagnosis and subsequent reporting (62). Furthermore, the absence of a significant association between aggregate vaccination coverage and notification rates at the district level likely reflects the limited variability in coverage (which was consistently high across districts) and the ecological nature of this analysis, which cannot account for individual-level protection. This ecological finding, which does not measure individual-level vaccine effectiveness and should not be construed as evidence against it, is consistent with the broader evidence that high childhood coverage alone cannot prevent resurgences (2).

Pertussis exhibited a consistent seasonal pattern in Chongqing, with annual peaks in May–June during 2013–2019, 2022, and 2024, but this pattern was disrupted in 2020–2021, likely due to COVID-19-related non-pharmaceutical interventions (NPIs) that suppressed respiratory pathogen transmission (63). This seasonal profile aligns with observations from other regions in China (18), France (8), Germany (64), and Vietnam (65). However, pertussis seasonality remains ambiguous: in winter and spring, high prevalence of influenza, mycoplasma, and respiratory syncytial virus infections leads to widespread cough symptoms, which overlap with those of atypical pertussis and complicate its diagnosis and reporting, potentially resulting in underreporting (8, 48). Pertussis cases are often misdiagnosed as other respiratory viral infections during winter, contributing to unclear seasonality (48).

Pertussis notification rates in Chongqing from 2013 to 2024 exhibited a non-random spatial distribution at the district/county level. Global spatial autocorrelation analysis confirmed a non-random, clustered distribution of pertussis cases across Chongqing. Spatial hotspots demonstrated both temporal dynamism and persistent concentration in the Central Urban Area and adjacent regions throughout the study period. These findings are consistent with the spatial patterns of pertussis notification rates, which were consistently higher in the Central Urban Area. This clustering may be attributed to the region’s role as the political and transportation hub of the municipality, supporting high population densities and substantial inbound migration (20, 66). Superior healthcare infrastructure, greater accuracy and sensitivity of surveillance systems, and well-developed transportation networks likely contribute to higher case detection rates in these areas (13). Furthermore, factors such as economic status, geographic location, and frequent population movement likely contribute to greater pertussis clustering in the Central Urban Area.

Spatiotemporal analysis further confirmed that the most likely cluster of pertussis cases was consistently located within the Central Urban Area, reinforcing its status as a high-risk region. This pattern is consistent with the typical diffusion of infectious diseases, which often emerge initially in urban centers before spreading to outlying areas (17, 67). The extreme RR estimates observed in some clusters should be interpreted with caution. While statistically significant, these implausibly high values are likely methodological artifacts arising from the analysis of areas with low expected case counts (33, 68, 69). Thus, statistical significance reliably identifies foci of elevated transmission, but the RR magnitude is best interpreted as a qualitative, rather than precise quantitative, indicator of elevated risk. The concentration of population, high-frequency human interaction (19, 20, 66), and superior transportation networks in this core area provide ideal conditions for pertussis transmission. Furthermore, the aggregation of advanced medical facilities likely enhances surveillance sensitivity through more frequent testing, potentially increasing the recorded notification rate in this area. The epidemiology of pertussis is shaped by a multifaceted confluence of these social and biological factors (2, 70).

Our study offers a comprehensive 12-year longitudinal analysis of pertussis epidemiology in Chongqing, one of China’s four major municipalities. To our knowledge, it represents one of the most detailed integrations of demographic evolution, spatiotemporal clustering, and vaccination coverage dynamics within a dense urban setting in China, providing empirical evidence that aligns with the national immunization schedule revision effective in 2025 (22). A key finding from this detailed analysis is the notable shift in age distribution: although infants <1 year were the most affected group between 2013 and 2023, school-aged children (5–10 years) emerged as the predominant group in 2024, bearing a burden 7.4 times that of infants. This clear transition—alongside the persistently high notification rate among very young infants—offers robust, evidence-based justification for accelerating the development and rollout of a maternal Tdap immunization program to protect newborns during early infancy.

This study has several limitations that should be considered when interpreting the findings. First, as with all analyses based on passive surveillance data, our observed pertussis notification rates are likely subject to significant underascertainment. Recent evidence from active surveillance studies in China demonstrates that community-based incidence estimates substantially exceed rates reported through passive systems, confirming that passive surveillance captures only a fraction of the true disease burden, particularly among older children and adults (71). The probability of case identification is influenced by multiple factors that exhibit geographic and age-related variation, including healthcare-seeking behavior, access to diagnostic services, and testing practices (71, 72). Second, although we employed age-stratified analyses to account for demographic changes, the absence of nationally representative standard population weights precluded calculation of age-standardized rates. This limitation is partially mitigated by our age-specific analytical approach, which enables valid comparisons within specific age groups.

In conclusion, our analysis revealed a substantial resurgence of pertussis in Chongqing from 2013 to 2024, characterized by a pivotal epidemiological shift in 2024, when school-aged children (5–10 years) became the largest case group while infants maintained the highest notification rates. This finding underscores the critical need to implement seroepidemiological surveillance to directly measure population immunity and rationally guide the development of targeted interventions.