We conducted a retrospective, observational study of pediatric patients within the VSD population to estimate RV, DTaP, and PCV vaccine coverage for infants born between January 1, 2018 and May 31, 2023, to evaluate associations between demographic and health-seeking factors and delays in vaccine administration. The VSD is a collaborative project between the Centers for Disease Control and Prevention’s (CDC’s) Immunization Safety Office and 13 health care organizations contributing data for approximately 4% of the US population annually.^20–22 Patients are included in the VSD cohort based on enrollment, defined by health insurance coverage, or for one VSD site empanelment, defined as having an ambulatory visit in the previous 12 months.²³ Thus, the population included in this study is primarily insured. This study was approved by the institutional review boards of all participating sites with a waiver of informed consent, and it followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline. This activity was reviewed by the CDC and is consistent with applicable federal law and CDC policy (see eg, 45 C.F.R. part 46.102[l][2], 21 C.F.R. part 56; 42 U.S.C. §241[d]; 5 U.S.C. §552a; 44 U.S.C. §3501 et seq).

The study population was infants aged 0 to 12 months who were born from January 1, 2018, through May 31, 2023. We required that infants have a minimum of 9 months of enrollment in the first 12 months of life with at least 1 medical visit between 9 and 12 months at any of the 8 participating VSD sites to improve capture of vaccinations administered during the first year of life. Participating VSD sites are located in California, Washington, Oregon, Colorado, Minnesota, and Wisconsin.

Vaccines were identified from standardized VSD files, including administrations recorded in the electronic health record (EHR), medical or pharmacy claims, or bidirectional communication with jurisdictional immunization information systems.²¹ We evaluated vaccine coverage for monthly birth cohorts at age 5 months and 12 months. At age 5 months, we included first and second doses of RV and 2 doses of DTaP and PCV vaccines for infants born January 1, 2018, through May 31, 2023. At age 12 months, we evaluated completion of recommended doses of RV, DTaP, and PCV for infants born January 1, 2018, through October 31, 2022. There are 2 currently licensed RV vaccines: Rotarix (RV1) is a 2-dose series and Rotateq (RV5) is a 3-dose series. For PCV, we only included PCV13 in this evaluation. We evaluated coverage for individual vaccines and the combined UTD status for the recommended number of vaccine doses by age (Supplemental Table 1). To be considered valid, we required that vaccine administrations occurred by at least 42 days of life and that vaccines from the same vaccine series had at least 28 days between doses.

We included infant age, sex, race and ethnicity, both inpatient and outpatient health use, and Medicaid obtained from standardized VSD files that capture data from EHRs and claims databases. Infant race and ethnicity were grouped into the following categories: non-Hispanic white (“White”), non-Hispanic Black or African American (“Black”), non-Hispanic Asian (“Asian”), Hispanic or Latino, and “Other,” which includes all other racial categories and missing race and ethnicity. Infant race and ethnicity were included to allow for assessment of racial and ethnic disparities and was based on data provided by the infant’s caregiver. We defined inpatient health use as the number and duration of inpatient encounters excluding birth hospitalizations: 0, at least 1 lasting 0 to 7 days, or at least 1 lasting more than 7 days. We defined outpatient health use as the number of outpatient primary care encounters (1–4 or ≥5 encounters) including both preventive care and acute care visits. Medicaid was defined as any Medicaid enrollment in the first year of life. We also included the infant’s caregiver’s primary language and need for interpreter. Primary language was categorized as English, Spanish, or non-English/non-Spanish. Need for interpreter was defined by whether a bilingual interpreter in English and caregiver’s primary language was requested.

We calculated vaccination coverage at 5 and 12 months by dividing the number of infants who received the selected vaccinations by the number of eligible infants in each monthly birth cohort, defined as all infants born in a calendar month during the study period. We estimated 95% CI using a generalized linear model with binomial distribution and logit link function. Coverage at 5 months included the study period of June 2018 to October 2023 and coverage at 12 months was assessed from January 2019 through October 2023. We stratified results by infant race and ethnicity, primary language, need for interpreter, and Medicaid for RV vaccines only.

We identified 4 distinct time periods of vaccine administrations related to the COVID-19 pandemic based on known disruptions to care^6,12 and posthoc trends in vaccination coverage: (1) prepandemic (ie, June 2018 through mid-March 2020), (2) age-limited preventive care (ie, mid-March through mid-May 2020), (3) expanded primary care (ie, mid-May 2020 through mid-June 2021), and (4) routine primary care (ie, mid-June 2021 through October 2023). We then assigned monthly birth cohorts to these time periods based on the expected timing of receiving first vaccine doses for 5-month coverage (ie, age 2 months) and second vaccine doses for 12-month coverage (ie, age 4 months). Monthly birth cohorts associated with the 4 time periods are shown in Supplemental Table 2.

We evaluated the association of the caregiver’s primary language or need for interpreter, infant race and ethnicity, Medicaid, health care use, and time period with vaccine uptake. This was done using time-to-event analysis in multi-variable Cox proportional hazards regression models adjusted for the previously mentioned covariates and for uptake by 12 months, delay of first dose past 15 weeks. For models estimating RV uptake or UTD status a covariate indicating RV product (RV1 or RV5) was included. We report adjusted hazard ratios (aHR) with 95% CI. We estimated aHRs for time to first RV vaccine censoring at 5 months and time to second RV vaccine censoring at 12 months, comparing infant’s racial and ethnic group and use of Medicaid or not by time period. We also evaluated the association between infant’s race and ethnicity and time to first or second RV vaccine stratified by Medicaid. Analysis was performed using SAS/STAT and SAS/GRAPH software (Version 9.4, SAS Institute Inc.).

We included 395 143 infants in 65 unique monthly birth cohorts: 48.4% female, 51.6% male, 14.7% non-Hispanic Asian (“Asian”), 5.8% non-Hispanic Black (“Black”), 34.2% Hispanic, and 31.8% non-Hispanic white (“White”) (Table 1). Spanish was reported as the primary language for 3.5% of caregivers, 2.0% reported a primary language other than English or Spanish, and 3.0% requested interpreter services at a medical visit. Medicaid was identified for 20.6%. Most (94.1%) infants had 5 or more primary care encounters in the first year of life, 6.1% of infants had 1 or more inpatient hospitalizations lasting 0 to 7 days, and 1.0% of infants had 1 or more inpatient hospitalization(s) lasting more than 7 days.

Vaccine coverage among infants by age 5 and 12 months is shown in Figure 1 and Supplemental Tables 3 and 4. Vaccine coverage at 5 months increased throughout the prepandemic period (Supplemental Table 3) and decreased during the pandemic. By the end of the evaluation period, October 2023, coverage for individual vaccines remained below prepandemic levels. UTD coverage at 5 months was 87.8% (95% CI 86.9–88.7) in February 2020 vs 80.8% (95% CI 79.8–81.8) in October 2023. Among infants aged 12 months across all time periods, coverage was highest for 3 doses of DTaP, followed by 3 doses of PCV, and then complete RV series. UTD coverage at 12 months was highest in January 2020 (92.3% [95% CI 91.6–92.9]) and decreased to 89.6% (95% CI 88.8–90.3) in October 2023 (Supplemental Table 4).

Demographic factors associated with lower vaccine uptake for all 3 vaccines among infants at both age 5 and 12 months included Black race compared with white (aHR 0.86 [95% CI 0.85–0.87] and 0.84 [95% CI 0.83–0.85]), caregiver’s primary language that was not English or Spanish (aHR 0.95 [95% CI 0.92–0.97] and 0.97 [95% CI 0.94–0.99]) compared with English, and Medicaid (aHR 0.73 [95% CI 0.73–0.74] and 0.82 [95% CI 0.81–0.82]) compared with no Medicaid (Table 2). There was higher vaccine uptake for Asian infants at 5 and 12 months compared with white infants with UTD status 1.32 (95% CI 1.30–1.33) and 1.27 (95% CI 1.25–1.28), respectively. Health care use was also associated with vaccine uptake. There was lower vaccine uptake among infants having 1 to 4 primary care visits in the first year of life compared with those having at least 5 visits at both 5 months (aHR 0.50 [95% CI 0.49–0.51]) and 12 months (aHR 0.53 [95% CI 0.52–0.54]). Any inpatient admission in the first year of life compared with no inpatient admissions was also associated with lower vaccine uptake. Finally, there was lower uptake after the COVID-19 pandemic began compared with the prepandemic time period with lowest uptake occurring during age-limited preventive care.

Throughout the study period, coverage for 2 doses of RV vaccine by 5 months and complete RV series by 12 months was consistently lower for Black infants compared with infants from other race and ethnic groups and for those with Medicaid compared with no Medicaid (Figure 2). There were no clear disparities in RV vaccine coverage noted for caregiver’s primary language or need for interpreter (Supplemental Figure 1). Compared with white infants, Asian and Hispanic infants had higher uptake of the first RV vaccine by 5 months and second RV vaccine by 12 months across time periods (Table 3).

The identified disparities in uptake of RV vaccine by racial and ethnic groups are modified by having Medicaid. RV vaccine uptake among Black infants with no Medicaid compared with white infants with no Medicaid was 0.92 [95% CI 0.90–0.94] at 5 months and 0.86 [95% CI 0.85–0.88] at 12 months. In Black infants with Medicaid compared with white infants with Medicaid associations were 1.03 [95% CI 1.00–1.06] at 5 months and 0.94 [95% CI 0.91–0.97] at 12 months. Different levels of uptake by Medicaid were also found among Hispanic infants compared with white infants with and without Medicaid. At 5 months, RV vaccine uptake among Hispanic infants with Medicaid was 1.18 [95% CI 1.16–1.20] and without Medicaid 1.03 [95% CI 1.02–1.04]. At 12 months associations were 1.19 [95% CI 1.16–1.21] with Medicaid and 1.05 [95% CI 1.04–1.06] without Medicaid.