Association of Maternal Tdap Recommendations With Pertussis Hospitalizations of Young Infants

We used the Kids’ Inpatient Database (KID), which is the largest all-payer pediatric inpatient database in the United States that yields national estimates of inpatient stays.⁶  KID data are produced every 3 to 4 years and represent a sample of hospitalizations. We included data from 2000 to 2016 that were weighted using the weights provided in KID to generate national estimates. Infants <1 month of age with a pertussis diagnosis were identified through International Classification of Diseases, Ninth Revision, Clinical Modification, and International Classification of Diseases, Tenth Revision, Clinical Modification, codes (033.x, 484.3, A37.xx). We selected age <1 month to exclude the potential effects of active immunization against pertussis among older infants, who may be vaccinated as early as 6 weeks of age.⁷  As maternal history was unavailable in KID, premature infants were excluded (using Feudtner’s complex chronic condition code for prematurity) because their mothers were less likely to have received the Tdap vaccine on the basis of optimal timing of administration.^2,8 

The monthly hospitalization rate was calculated as the number of pertussis hospitalizations per 100 000 term births as reported by the National Center for Health Statistics.⁹  Interrupted time series with weighted Poisson regression modeling was used to evaluate differences between hospitalization rates before (January 2000–October 2012) and after (November 2012–December 2016) guideline implementation. Length of stay (LOS) in days and mortality rates were modeled using weighted Poisson regression and weighted Firth-based logistic regression, respectively. Model covariates included sex, race, payer, region, and presence of a complex chronic condition.

Analyses were performed using R version 3.6.2 statistical software (R Foundation for Statistical Computing). This study was designated as nonhuman subjects research by the institutional review board.

We identified 958 infants with a pertussis diagnosis (Table 1). The overall rate of pertussis hospitalizations was 5.06 per 100 000 infants (95% confidence interval [CI], 4.36–5.76) before the Tdap vaccination recommendation and 2.15 per 100 000 infants (95% CI, 1.49–2.81) after (Fig 1). After adjusting for time and other covariates in the interrupted time series, this level of change in hospitalization rate was significant (rate ratio, 0.24; 95% CI, 0.15–0.38; P < .001).

There was no significant difference in LOS for pertussis hospitalization before versus after the Tdap vaccination recommendation (median, 6 days [interquartile range, 3–10] vs. 6 days [interquartile range, 3–12]; P = .99). Death occurred in 2.3% of infants, and mortality rates did not change after guideline implementation (adjusted odds ratio, 0.63; 95% CI, 0.39–1.01; P > .99).

Our findings reveal a sharp decline in pertussis hospitalizations among young infants after the recommendation for antenatal Tdap vaccination. These results are complemented by those of other studies, and^4,5  the first to describe this decline using a nationally representative population of hospitalized infants.

Several limitations are worth noting. Given the cyclic nature of pertussis, it is possible that the decrease in hospitalizations we observed may have overlapped with epidemiological fluctuations in pertussis. If this were the case, we could expect similar patterns among other ages as well, yet both the proportion and the incidence of hospitalizations among adults increased during the study period.¹  In turn, this contrast is supported by the contribution of maternal vaccination to the decrease in hospitalizations among young infants in our study. Importantly, the data available after the 2012 recommendation were limited to a short time period, and evaluation of long-term trends in hospitalization is warranted once more contemporary data become available.

It was difficult to conclude whether significant differences in death occurred over time due to the limited number of cases. Additionally, no significant differences were found in LOS after guideline implementation, and it is plausible that these hospitalizations were among infants of mothers who did not receive the Tdap vaccine. Because Tdap vaccination history was unavailable, it was not possible to directly link vaccination status with hospitalization. However, antenatal Tdap vaccination is likely a significant contributor to our findings given its role in prevention and reduced severity of disease compared with infants of unvaccinated mothers.^4,10,11  Furthermore, vaccination of the mother also indirectly protects the infant because exposure to pertussis at such a young age is likely from a household contact.

It is essential to protect the vulnerable population of young infants from pertussis-related morbidity and mortality, and our findings in this study support the importance of promoting maternal Tdap vaccination.