Addressing Disparities In Pertussis Epidemiology

Pertussis is the most poorly controlled bacterial vaccine-preventable disease in the United States and causes a substantial burden of morbidity and mortality among young children. The dramatic increase in pertussis incidence in recent years, both on national and local levels, underscores the need for more effective prevention and control strategies. For several decades, significant sociodemographic disparities in pertussis risk, morbidity and mortality have been observed, yet reasons to explain these disparities are poorly understood. Given well-established relationships between health and place, understanding the role of neighborhood-level exposures associated with pertussis may provide insights into the mechanisms underlying observed disparities. Further, continual refinement of immunization programs is critical for optimal prevention and control. By leveraging administrative data from a variety of sources at the Philadelphia Department of Public Health to create a population-based cohort of Philadelphia children, this dissertation presents three retrospective cohort studies that provide a comprehensive investigation of individual-level, neighborhood-level, and cross-level effects on pertussis risk and delayed pertussis vaccination in order to effectively refine targeted prevention efforts and narrow pertussis-related disparities. In the first study, we determine the sociodemographic and neighborhood-level characteristics associated with pertussis. In the second study, we determine the sociodemographic and neighborhood-level characteristics associated with delayed initiation and completion of the primary pertussis vaccine series, as well as time to first pertussis immunization. These two studies present multilevel analyses utilizing generalized estimating equations accounting for correlated neighborhood-level errors to estimate the adjusted odds of pertussis and age-appropriate vaccination, respectively. The second study also uses Cox Proportional Hazards models with adjusted standard errors to estimate hazards of delayed initiation of pertussis immunization. Finally, in the third study, we evaluate the impact of an immunization program outreach policy change to target young infants not up-to-date with the recommended vaccine series on delayed pertussis immunization and sociodemographic disparities. We utilize population-level interrupted time series analyses to understand the impacts of the policy change. Overall, this dissertation advances our understanding of the role of neighborhood-level exposures associated with pertussis and delayed pertussis vaccination and identifies actionable opportunities to bolster pertussis prevention efforts in Philadelphia