Infant Respiratory Syncytial Virus Immunization Coverage in the Vaccine Safety Datalink: 2023–2024 Free

Respiratory syncytial virus (RSV) is a significant cause of respiratory illness in young children and the leading cause of hospitalization in infants.^1–3 In 2023, the Advisory Committee for Immunization Practices (ACIP) recommended 2 products for the prevention of RSV-associated respiratory disease in infants: a bivalent RSV prefusion F protein-based vaccine (Abrysvo, Pfizer Inc) for administration during 32 through 36 weeks of pregnancy and the long-acting monoclonal antibody nirsevimab (Beyfortus, Sanofi, and AstraZeneca) for administration to infants after birth.^4,5 Each product is recommended for infant protection, but both are not needed unless the infant was born within 14 days of the antenatal vaccination, or in special circumstances when additional administration of nirsevimab may provide increased protection to the infant per the clinical judgement of the health care provider. Since these recommendations, few robust estimates of product uptake and infant RSV immunization coverage in the United States have been published. Coverage studies provide information about the real-world implementation of vaccine recommendations and can identify nonadherence to recommendations and disparities in immunization.

To help fill this gap in the literature, we used the Vaccine Safety Datalink (VSD) population to assess infant RSV immunization through antenatal RSV vaccination or receipt of nirsevimab among linked pregnancy-infant dyads.

Within a defined cohort of pregnant women and their linked infants across multiple study sites, we performed a descriptive study examining infant RSV immunization, defined as exposure to antenatal RSV vaccination 14 days or more before birth or receipt of nirsevimab after birth.

The VSD was established in 1990 as a collaboration between the Centers for Disease Control and Prevention (CDC) and several health care organizations (sites); the current network includes 11 data-contributing sites.⁶ The VSD assesses population-based vaccine safety questions and examines vaccine uptake and coverage using electronic health record (EHR) data from participating sites.⁷ Ten VSD sites contributed data to this study: Denver Health, Harvard Pilgrim Healthcare Institute, HealthPartners, Marshfield Clinic, and the Colorado, Mid-Atlantic States, Northern California, Northwest, Southern California, and Washington regions of Kaiser Permanente. The demographic characteristics of those enrolled in these health care organizations have been found to be representative of the geographic areas covered by the health systems and the general US population, but those analyses were not restricted to pregnant populations.^8,9 

Linked pregnancies and infants (dyads) were identified within the VSD data, as described below. A validated algorithm was used to identify pregnancies on a weekly basis by applying a hierarchical approach using the International Classification of Diseases, 10th Revision,Clinical Modification (ICD-10-CM) diagnosis and procedure codes indicating a pregnancy (eg, pregnancy outcome, prenatal care encounter).¹⁰ Additional data, including estimated date of delivery, last menstrual period, gestational age diagnoses, and trimester-specific care diagnoses, were used to assign gestational age.¹⁰ Pregnancies eligible for inclusion in the analysis reached at least 32 weeks’ gestation between September 22, 2023, and January 31, 2024, and ended in a live birth before April 1, 2024. Included pregnant women were ages 12 to 55 years at the start of the pregnancy and continuously enrolled in the VSD health system for the duration of the pregnancy and throughout the September 22 to January 31 period to maximize RSV vaccine data capture.

Infants were identified from the pregnancies described above using a VSD site-specific pregnancy-baby linkage process that uses delivery records, patient account information, patient address, and delivery and birth dates. Eligible infants were born between September 22, 2023, and March 31, 2024, and were enrolled at their respective VSD health system for 30 days or more within their first 90 days of life.

This study was approved by the institutional review boards of all participating sites with a waiver of informed consent because it was a minimal risk observational study and was conducted consistent with federal law and CDC policy (see 45 C.F.R. part 46.114; 21 C.F.R. part 56.114).

Antenatal vaccine receipt was determined using EHR, claims, and bidirectional communication with jurisdictional immunization registries to identify vaccines administered within and outside the health care system.¹¹ RSV vaccines were identified using CVX codes 303–305 and 314. All eligible RSV vaccines administered during the pregnancy were included in analyses; very few antenatal RSV vaccines were administered at participating sites after January 31. Monoclonal antibody therapy (nirsevimab) was identified from a standardized VSD file containing data on nirsevimab administered in both the inpatient and outpatient settings within each participating health system. To create this file, nirsevimab data were extracted from immunization and pharmacy files, and jurisdictional immunization information systems using text and CVX code searches (CVX codes 314, 315). All nirsevimab doses administered within the eligible infant population through March 31, 2024, were included in analyses.

The timing of antenatal RSV vaccination was examined by gestational age (<32 weeks, 32–36 weeks, ≥37 weeks). In addition, RSV vaccines administered fewer than 14 days prior to birth were flagged, and receipt of nirsevimab within this group of infants was explored. Age at nirsevimab receipt was collected starting at the day of birth (day 0) and in subsequent 1-week increments to examine patterns of administration, both overall and stratified by month of birth.

Infant RSV immunization was defined as exposure to antenatal RSV vaccination 14 days or more before birth or receipt of nirsevimab after birth. Infants exposed to RSV vaccination fewer than 14 days before birth who did not receive nirsevimab after birth were considered unimmunized in analyses. Infants exposed to RSV vaccination fewer than 14 days before birth who received nirsevimab after birth were considered immunized via nirsevimab, exclusively, in analyses. Crude RSV immunization coverage was calculated by dividing the number of immunized infants by the total number of eligible infants identified. Immunization coverage was calculated overall for infants born during the September 2023 to March 2024 study period and by infant month of birth, maternal race and ethnicity (non-Hispanic [NH] American Indian or Alaskan Native, NH Asian, NH Black, Hispanic or Latino, NH Middle Eastern or North African [MENA], NH another race or multiple races, NH Native Hawaiian or other Pacific Islander, unknown or missing race, or NH white, as documented in the EHR); age group (12–17, 18–24, 25–34, 35–49, 50–55 years as of pregnancy start); and health care utilization (number of in-person clinical encounters with any diagnosis during the pregnancy, measured categorically as <10, 10–19, 20–29, and ≥30). Inclusion of race and ethnicity in analyses was critical to evaluate potential disparities across populations. Ninety-five percent Wald confidence intervals were calculated for immunization coverage estimates.

A total of 43 722 eligible pregnancies were identified during the study period. Eighty-three percent of eligible pregnancies were linked to 1 eligible infant or more, resulting in 36 949 pregnancy-infant dyads included in RSV immunization coverage estimates (Figure 1, Supplemental Table 1).

Descriptive characteristics of the dyads are shown in Table 1. The median number of eligible infants born per study month (September 2023 through March 2024) was 6502, with few September births given the September 22 study start, and a high of 6958 infants born in December. Most pregnant women were aged 25 to 34 years (63.3%), and 90.7% were between the ages of 25 and 49 years. The 3 most commonly documented maternal races and ethnicities were Hispanic or Latino (34.5%), NH white (29.9%), and NH Asian (17.8%). Two-thirds of pregnant women had 10 to 19 health care visits during pregnancy.

Dates of RSV immunization product usage across VSD sites are shown in Supplemental Table 2. A total of 11 774 of the 36 949 linked infants were exposed to antenatal RSV vaccine (31.9%; Table 2). Of these, 810 antenatal RSV vaccines (6.9%) were administered fewer than 14 days before birth; this included receipt at any gestational age. Among the 10 964 antenatal RSV vaccines administered 14 days or more before birth, 95% were given within 32 to 36 weeks’ gestation, the window recommended by the ACIP. A total of 15 708 linked infants received nirsevimab (42.5%; Table 2). Among those, 9146 infants (58.2%) received nirsevimab in the first week of life and 2511 infants (16.0%) received nirsevimab at 28 days of age or older (data not shown). Nirsevimab administration after the first week of life was less frequent for infants born between January and March 2024 (Supplemental Figure 1). Of the 810 infants exposed to antenatal RSV vaccine fewer than 14 days before birth, 583 (72.0%) subsequently received nirsevimab, as recommended by ACIP. Across the study population, only 174 infants (0.5%) were exposed to antenatal RSV vaccine 14 days or more before birth and later received nirsevimab (dual exposure; data not shown). Crude infant RSV immunization coverage was 71.7% (95% CI, 71.3%–72.2%).

By month of birth, RSV immunization coverage ranged from 58.7% (95% CI, 51.9%–65.5%) in September 2023 to 77.5% (95% CI, 76.5%–78.5%) in January 2024, with the distribution of RSV product shifting over the study period (Figure 2). Infants born September through December 2023 more often received nirsevimab, whereas those born January through March 2024 were more often exposed to antenatal RSV vaccination.

RSV immunization coverage by race and ethnicity ranged from 60.2% (95% CI, 55.3%–65.0%) among infants born to NH MENA mothers and 60.5% among infants born to NH Black mothers (95% CI, 58.7%–62.3%) to 83.7% (95% CI, 82.8%–84.6%) among infants born to NH Asian mothers, with confidence intervals greater than or equal to 10 percentage points in 3 racial and ethnic groups (Figure 3). Nirsevimab accounted for a larger proportion of RSV immunization compared with antenatal RSV vaccination in all racial and ethnic groups.

By age group, RSV immunization coverage ranged from 63.6% (95% CI, 35.2%–92.1%) among infants born to mothers aged 50 to 55 years to 81.4% (95% CI, 74.3%–88.4%) among infants born to mothers aged 12 to 17 years. Aside from the 12 to 17 and 50 to 55-year age groups, which had small numbers and wide confidence intervals, RSV immunization coverage increased with the age of the mother (Figure 4). The distribution of the RSV immunization product varied across age groups, with 67.8% of infants born to 12 to 17-year-olds receiving nirsevimab, compared with 18.2% of infants born to 50 to 55-year-olds. The percent of infants protected by antenatal RSV vaccination, specifically, increased with each increasing age group (Figure 4). There was little variability in RSV immunization coverage by health care utilization during the pregnancy; coverage ranged from 65.3% (<10 visits) to 72.5% (both 10–19 and 20–29 visits, data not shown).

In the first respiratory season following the licensure and recommendation of an antenatal RSV vaccine and the long-acting monoclonal antibody, nirsevimab, we observed 72% coverage with either product among infants born in 10 VSD health care systems. Both products were widely used in this population, but product-specific coverage varied by birth month, with infants born in the fall more likely to receive nirsevimab and infants born in later months more likely to be exposed to antenatal RSV vaccination. Disparities in coverage were observed by race and ethnicity with the highest estimates among infants born to NH Asian mothers and the lowest estimates among infants born to NH Black and NH MENA mothers.

Despite the challenges posed by the roll-out of 2 novel products¹² as well as shortages and delays in the 2023 to 2024 season,¹³ participating VSD sites were able to achieve high RSV immunization coverage. Apart from one health care system that provided only nirsevimab, both products were offered at all other participating VSD sites, and providers were able to administer nirsevimab at delivery hospitalizations or well-child visits to infants who were born before antenatal vaccination was possible or whose mother was unvaccinated. However, nirsevimab shortages left some unvaccinated infants born in September and October without protection for weeks to months because they were not able to receive the product at birth.^13,14 Further evaluations of the timing of RSV immunization are warranted for future seasons, when shortages and initial roll-out challenges have been alleviated.

Few coverage estimates for antenatal RSV vaccine or nirsevimab in the United States are available for comparison with our findings. In an internet panel survey conducted by the CDC in April 2024, of 678 eligible pregnant women, 32.6% reported receipt of the antenatal RSV vaccine. Among 866 women surveyed with an infant born between August 2023 and March 2024, 44.6% reported receipt of nirsevimab by the infant. The survey found an overall RSV immunization coverage of 55.8% among infants.¹⁵ The Wisconsin state health department reported that 17% of birthing parents received 2023 to 2024 RSV vaccines and 18% of infants born between October 1, 2023, and March 31, 2024, received nirsevimab, for a combined infant RSV immunization estimate of 36%. Similar to our findings, within the Wisconsin data, NH Black mothers had the lowest uptake of either RSV product; MENA was not specifically included in that race and ethnicity analysis.¹⁶ Both the survey estimates and the Wisconsin estimates are lower than findings in the VSD, likely due to population differences, pregnancy and exposure ascertainment methods, and the health care system enrollment criteria required for inclusion in our analysis. A single-site examination of infants born within Kaiser Permanente Northern California (a VSD site) reported similar, but slightly higher, estimates than the VSD-wide results described here, with 78% of infants protected by either of the RSV products.¹⁷ The VSD coverage estimates are also similar to findings from RSV immunization intention surveys, which reported that 54% to 71% of pregnant women were likely to receive the RSV vaccine and that 89% of pregnant women intended for their infant to be immunized against RSV by 1 of the 2 available products.^18–20 

Despite high RSV immunization coverage in our study population overall, notable disparities in estimates by race and ethnicity were observed. Similar to other VSD and national coverage studies for influenza, Tdap, and COVID-19 vaccination during pregnancy, the lowest coverage levels were observed in NH Black mothers.^21–24 We saw similarly low coverage among NH MENA mothers and their infants, a group that is underrepresented in the literature. However, even in our large VSD population, the MENA group was small; therefore, our findings should be interpreted with care. Like findings from studies of other vaccines administered during pregnancy, we found the highest coverage among NH Asian mothers and their infants; overall RSV immunization coverage was more than 23 percentage points higher among NH Asian dyads compared with the coverage among NH Black and NH MENA dyads. Interestingly, we also found high immunization coverage among infants born to NH American Indian and Alaskan Native mothers, with notably high uptake of nirsevimab in their infants. Although these findings are encouraging, the small population size should be noted. The VSD will continue to monitor infant RSV immunization coverage in the upcoming 2024 to 2025 season, including the examination of coverage by race and ethnicity.

In addition to the high RSV immunization coverage identified within this VSD population, we found high adherence to vaccination recommendations and clinical guidance among vaccine providers. Our findings suggest that providers were closely following the ACIP recommendations to administer RSV vaccination during 32 to 36 weeks’ gestation;⁴ very few doses were administered outside of this window. Providers were also able to identify infants who were not protected by passive antibody transfer because they were born within 14 days of antenatal RSV vaccination; 72% of these infants subsequently received nirsevimab. One-hundred seventy-four infants were exposed to both antenatal RSV vaccination 14 days or more before birth and nirsevimab; it is unknown why these infants received nirsevimab. It is possible that information about antenatal vaccination was unavailable to the pediatric provider. It is also possible that these infants were medically fragile and/or had a medical condition that warranted receipt of both products.⁵ Additional investigation of the infants’ medical records would be needed to describe the reasons for receipt of both products.

Our analysis is among the first to examine infant RSV immunization, with comprehensive capture of both antenatal RSV vaccination and nirsevimab receipt in a large, well-defined population. The VSD health system memberships that provided the diverse denominator for this large study have been shown to be representative of both their geographic regions and the United States.^8,9 The long-established VSD data sources used here include robust EHR systems and are linked to jurisdictional immunization information systems at most sites,^7,11 and the VSD enrollment requirement for the current analysis provides confidence that capture of RSV product administration is accurate and comprehensive. Additionally, exposure information is taken directly from EHR data, minimizing the recall bias that can affect coverage estimates using survey data.

Our study has limitations. Although the general VSD population is demographically representative, it may not be representative of the broader pregnant population by race and ethnicity. Additionally, although VSD health systems serve uninsured and underinsured patients, the VSD is a primarily privately insured population and the large VSD health systems may have been better able to obtain and administer both RSV products than other health care delivery sites, such as private practices, who may have been more heavily impacted by the fall 2023 nirsevimab shortage. Our study inclusion criteria, although important for product ascertainment, increased the proportion of the study population with medical insurance and access to medical care, further limiting comparability with other study populations and generalizability with the larger US population. Second, our study question was specific to infants eligible for either RSV vaccination or nirsevimab receipt and did not examine infants born prior to RSV vaccine availability who were eligible for immunization via nirsevimab only (eg, infants <8 months of age as of September 21, 2023). RSV immunization in these older infants is an important study question and is being examined separately within the VSD. Third, the subset of our pregnancies that could not be linked to an eligible infant differed by age, and we are unable to examine the coverage in the unlinked group; that estimate may differ from our findings. The final study population included lower proportions of younger birthing parents (12–24 years) compared with the original pregnancies identified. Fourth, although our robust data sources and inclusion criteria should have maximized capture of RSV immunization products, it is possible that ascertainment is incomplete. However, this would have led to an underestimation of immunization coverage. Finally, although the VSD’s validated pregnancy algorithm has high accuracy, there is some minor imprecision around pregnancy start dates.¹⁰ This could have impacted results specific to gestational week but would not have impacted coverage estimates.

Ten US health systems included in the VSD experienced high infant RSV immunization coverage in the first season where an antenatal RSV vaccine and nirsevimab were available and recommended, even with limited product availability and a quick product implementation timeline following recommendations. Despite these successes, more than a quarter of infants enrolled in these health care systems remained unprotected from severe RSV, including 40% of infants born to NH Black and NH MENA mothers. RSV product usage and coverage disparities should be explored in other populations.

The authors thank the Vaccine Safety Datalink Project Managers, Data Managers, and Programmers from Denver Health, Harvard Pilgrim Research Institute, HealthPartners Institute, Kaiser Permanente Center for Health Research, Kaiser Permanente Department of Research and Evaluation, Kaiser Permanente Institute for Health Research, Kaiser Permanente Mid-Atlantic States, Kaiser Permanente Vaccine Study Center, Kaiser Permanente Washington Health Research Institute, and Marshfield Clinic Research Institute for their support of this work.