1607. Temporal Patterns and Spatial Synchrony in Pertussis Incidence—the United States, 2000–2017

BACKGROUND: Pertussis is a highly contagious, vaccine-preventable respiratory disease. Historically, pertussis incidence was cyclic with peaks in disease every 3–5 years. In the United States, reported pertussis has increased over the past few decades despite high vaccination coverage; however, there is currently no clear national spatiotemporal pattern. We aimed to assess whether geographically distinct areas in the United States: (1) share similar temporal patterns (trend and periodicity), and (2) were synchronous in peaks in pertussis incidence. METHODS: We used pertussis cases reported through the National Notifiable Diseases Surveillance System, and CDC Wonder bridged-race county population estimates, for 2000–2017. County-level pertussis case counts were aggregated by month, and counties were aggregated into population areas of ≥300,000 persons. For each population area, trend and dominant periods across the study period were extracted using wavelet analysis. Common temporal patterns were identified using hierarchical cluster analysis of trend and periodicity. Synchrony between population area pairs, and between each area and the country as a whole, were assessed using wavelet coherence and phase difference. RESULTS: There was substantial variability in temporal patterns, though geographically distinct population areas clustered by trend and similar dominant periods of 8 to <19 months, 19 to <38 months, and 38 to 71 months, with three main patterns accounting for 79% (400/506) of all population areas (Figures 1 and 2). The majority of areas had a background period of at least 38 months, and 87% (439/506) of population areas experienced a positive trend. However, only 37% (185/506) of areas were synchronous with the national time series at any time during 2000–2017. CONCLUSION: Spatiotemporal patterns in pertussis incidence are complex, and are heterogeneous across the United States. Although a background period of at least 38 months was identified in the majority of areas, similar to the historic perception of a 3–5-year cycle, higher frequency components were also identified. A better understanding of the current spatiotemporal patterns of pertussis will allow us to better characterize current epidemiology and improve prediction of future outbreaks. [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.