TL;DR
Pertussis (whooping cough) is making a comeback in Shandong, China, with incidence rates skyrocketing by 16.8 times between 2009 and 2017. This study reveals a concerning shift in both the timing and location of outbreaks, suggesting that current prevention strategies may need recalibration to address these evolving spatial-temporal dynamics.
What Problem Does This Paper Address?
Despite widespread vaccination programs, pertussis remains a persistent global health concern, with recent resurgences observed in several countries. In China, particularly in Shandong Province, there's been a dramatic increase in pertussis cases over the last decade. This rise is concerning, especially given reports of macrolide resistance in Bordetella pertussis and potentially inadequate population immunity. Understanding the spatial and temporal patterns of this resurgence is crucial for developing effective, targeted prevention and control strategies. The paper aims to fill this knowledge gap by analyzing how pertussis transmission has evolved in Shandong Province between 2009 and 2017.
Methods
The researchers employed a multi-faceted approach to dissect the spatial and temporal dynamics of pertussis in Shandong:
- Time Series Decomposition: This standard technique was used to break down weekly pertussis data into trend, seasonal, and residual components, allowing for the identification of seasonal patterns and their changes over time.
- Periodical Subsetting: The study period (2009-2017) was divided into three distinct three-year intervals (2009-2011, 2012-2014, 2015-2017) to compare how seasonality and spatial patterns evolved across these phases.
- Space-Time Cluster Analysis (SaTScan): This statistical software was used to identify geographical areas with a significantly higher-than-expected number of pertussis cases, pinpointing "hotspots" of infection.
- Logistic Regression Analysis: This method was applied to investigate the relationship between geographical coordinates (longitude and latitude) and the odds of increased pertussis incidence or spread between different time periods.
- Regression Tree Model: This non-parametric modeling approach was used to identify key geographical thresholds (like longitude) that significantly influenced changes in pertussis incidence between the early and late periods of the study.
Datasets
The study utilized weekly pertussis case data from the Chinese National Notifiable Disease Reporting System (CNNDRS) for the period of January 2009 to December 2017. Annual population data at the county level was obtained from the Shandong Provincial Bureau of Statistics.
Key Insights and Ideas
- Dramatic Increase: The sheer magnitude of the pertussis resurgence is striking, with a 16.8-fold increase in yearly incidence rates over less than a decade. This suggests that existing control measures might be losing effectiveness or that new factors are driving the epidemic.
- Shifting Seasonality: Pertussis outbreaks are not static. While early in the study period (2009-2011) a summer peak was observed, by the later period (2015-2017), the peak shifted to late summer and early autumn. This temporal shift is critical for timing public health interventions.
- Geographic Migration of Hotspots: The spatial distribution of high-risk areas also changed. Initially concentrated in southern Shandong (Linyi), the epicenters of pertussis transmission moved northward, particularly to the capital city, Jinan, and surrounding areas. This indicates a dynamic spread that needs continuous monitoring.
- Longitude as a Key Factor: The analysis revealed that longitude, specifically the 118.0°E meridian, acts as a significant dividing line. Counties west of this line experienced a disproportionately larger increase in pertussis incidence, suggesting a directional trend in the disease's spread.
- Eastward and Northward Expansion: Beyond the primary hotspots, there's evidence of pertussis spreading towards the east and north of the province, indicating a broader geographic expansion of the disease.
Key Results
- Incidence Rate Increase: Yearly pertussis incidence rates increased by 16.8 times from 2009 to 2017.
- Peak Shift: The peak incidence shifted from June-July in earlier periods to August-September in later periods.
- Spatial Shift: The primary high-risk areas shifted from Linyi in the south to Jinan in the northwest.
- Increased Relative Risk (RR): The relative risk of infection in identified clusters increased from 4.1 (95% CI: 2.3-7.4) in 2009-2011 to 6.1 (95% CI: 5.6-6.7) in 2015-2017.
- Odds of Increased Infection: West Shandong had significantly higher odds of increased infections (OR: 1.52, 95% CI: 1.05–2.17). Pertussis also showed increased odds of spreading to the east (OR: 2.32, 95% CI: 1.63–3.31) and north (OR: 1.69, 95% CI: 1.06–2.99) over time.
- Regression Tree Finding: The mean difference in yearly average pertussis incidence between 2009-2011 and 2015-2017 increased by more than 4-fold for counties with longitudes less than 118.0°E.
Limitations and Future Work
The authors acknowledge several limitations:
- Underreporting: The CNNDRS data only includes reported cases, potentially missing individuals who don't seek medical attention or are misdiagnosed.
- Reporting Bias: Increased physician awareness and reporting can influence case numbers.
- Diagnostic Shifts: Changes in diagnostic testing methods (e.g., the shift to PCR) could affect case detection.
Future work should focus on:
- Investigating the specific reasons for the increasing trend in pertussis, including potential factors like vaccine type changes, diagnostic test evolution, and the emergence of antibiotic resistance.
- Further exploring the reasons behind the changing seasonal patterns.
- Developing and optimizing preventative measures and control programs tailored to the identified high-risk spatial-temporal windows.
- Understanding the potential for international spread given the observed geographic expansion.
Summary updated