We conducted a longitudinal cohort study within the Vaccine Safety Datalink (VSD), a collaboration between the Centers for Disease Control and Prevention and multiple managed care organizations (MCOs) in the United States.¹² This study included 6 MCOs: Kaiser Permanente Washington, Kaiser Permanente Northwest, Kaiser Permanente Northern California, Kaiser Permanente Southern California, Kaiser Permanente Colorado, and the Marshfield Clinic. Managed care organization members were eligible to enter the study population when they met all the following criteria:

Cohort members were followed for 42 days¹³ from the date of their second tetanus-containing vaccine, or until they reached 65 years of age, disenrolled from the MCO, or died, whichever came first. We did not assess risks after a third dose of tetanus-containing vaccine, as this was rarely observed. We excluded MCO enrollees if they were pregnant during their subsequent dose of tetanus-containing vaccine, as rates and detection of adverse events may differ during pregnancy than during other times.

We identified vaccine type and date of receipt of tetanus-containing vaccines by using immunization databases at each MCO. These databases include all vaccinations given to enrollees at each MCO. The immunization databases at Kaiser Permanente Washington, Marshfield Clinic, and Kaiser Permanente Northwest also exchange data with state immunization information systems, allowing us to identify initial Tdap vaccines received prior to MCO enrollment or outside the MCO health-care system.

Consistent with a prior VSD study of Tdap safety,¹³ we considered 7 serious potential adverse events: seizure; cranial nerve disorders; limb swelling; pain in limb; cellulitis; paralytic syndromes; and encephalopathy, encephalitis, or meningitis. We did not include Guillain-Barré syndrome as an outcome,^13,14 as preliminary analyses suggested that we would detect fewer than 5 Guillain-Barré syndrome cases in our population. We identified potential adverse events by using International Classification of Diseases, version 9, Clinical Modification (ICD-9-CM) codes assigned to inpatient, outpatient, and emergency department encounters (Table 1). For each subject, we identified first diagnosis date of each outcome occurring within the observation period after their second tetanus toxoid-containing vaccination. To exclude events which may have had onset prior to vaccination, we excluded any codes during the postvaccination observation period if those specific codes also occurred during the 30 days prior to vaccination.

We used enrollment and medical utilization databases at each MCO to define demographics, MCO enrollment history, indicators of healthcare utilization (numbers of outpatient visits, well care visits, and hospitalizations in the past year), and receipt of other recommended vaccinations. We defined these covariates as of the dates of initial Tdap and of second tetanus-containing vaccine.

We compared covariate distributions between participants whose second dose of tetanus-containing vaccine was Tdap vs Td. We estimated rates of potential adverse events, with 95% CIs, assuming a Poisson distribution. To compare adverse event rates following Tdap vs Td as the second dose of tetanus-containing vaccine, we estimated rate ratios for each outcome of interest by using Poisson regression. A priori, models were adjusted for age at first receipt of Tdap, number of years between first Tdap and subsequent dose of tetanus-containing vaccine, and year of second dose of tetanus-containing vaccine. We further included any covariates that altered the rate ratios from this base model by 10% or more. As secondary analyses, we stratified by time between vaccine doses (<3 years vs ≥3 years after Tdap, the median interval between doses). Due to the smaller number of outcomes, these analyses were only adjusted for age at first Tdap, years between doses, and year of second dose. All analyses were conducted by using SAS version 9.4 (SAS Institute, Cary NC).

Unadjusted adverse event rates were lower following a second dose of Tdap than following a subsequent Td for cellulitis (5.2 per 10 000 vaccinees after Tdap vs 5.3 after Td), limb swelling (3.4 vs 9.3 per 10 000), pain in limb (16.9 vs 33.2 per 10 000), and seizure (8.3 vs 13.3 per 10 000). Cranial nerve disorders were slightly more common following Tdap (5.4 per 10 000) compared with Td (4.0 per 10 000). In adjusted analyses, receipt of Tdap vs Td was not significantly associated with any of these outcomes (Figure 1). Only 6 cases of encephalopathy/encephalitis/meningitis were detected (5 following Tdap and 1 following Td), precluding adjusted analyses, but the crude rate ratio (0.6) was not statistically significant (95% confidence interval [CI], 0.1 to 5.2). Similarly, only 23 cases of paralytic syndromes were detected, with incidence of 2.9 per 10 000 vaccinees for Tdap and 6.6 per 10 000 for Td (crude rate ratio 0.4, 95% CI, 0.2-1.2).

Among subjects receiving a second dose of tetanus-containing vaccine <3 years after initial Tdap, receipt of Tdap vs Td was not significantly associated with any study outcome (Figure 1). Among subjects receiving a second dose ≥3 years after initial Tdap, limb swelling and pain in limb were significantly less common among recipients of Tdap vs Td (Figure 1); significant associations were not observed for cellulitis, seizure, or cranial nerve disorders.