Vaccine Effectiveness Studies Research Articles

P-2046. Integrating Immunization Information System (IIS) and commercial health insurance claims data to improve COVID-19 vaccination capture among pregnant people

Abstract Background Several COVID-19 vaccine safety and effectiveness studies have used real-world data (RWD), including health insurance claims to delineate people’s vaccination history. However, COVID-19 vaccine administration is not completely captured in claims because vaccines administered in some settings, such as mass vaccination centers, do not generate insurance claims. Failure to capture complete vaccination history in claims data may result in misclassification of vaccination status leading to biased effectiveness and safety estimates. As part of a study to evaluate COVID-19 vaccine effectiveness in pregnant persons, we examined how linking claims with Immunization Information System (IIS) records improves vaccination record capture for pregnant people.Figure 1:Vaccination capture in claims and IIS Methods We identified individuals in a large commercial claims database who were pregnant between 12/11/2020 and 7/31/2023 and resided in 21 jurisdictions for which we had permission to use IIS data for public health purposes. We linked COVID-19 vaccination records from IIS and insurance claims and calculated the percentage of records that were captured in IIS only, claims only, and in both. We also compared COVID-19 vaccination capture in claims and IIS over time to assess how the completeness of IIS and claims data have changed over time. Results This study included 101,355 pregnant people, of which 13% (12,726) had ≥1 COVID-19 vaccination record only in IIS, 18% (17,787) had ≥1 vaccination record only in the claims data, and 36% (36,209) had vaccination records in both (Figure 1). Approximately 34% (34,633) did not have a vaccination record in IIS or claims. Over time, the percentages of individuals with vaccination records in IIS, claims, and both changed; the temporal patterns aligned with a shift from vaccinations at mass vaccination centers to pharmacies and clinics. Conclusion This study demonstrates the importance of integrating data from multiple RWD sources to capture vaccine history for vaccine effectiveness studies in pregnant people and the general population. Relying on only one data source may result in substantial misclassification of vaccinated individuals as unvaccinated, leading to biased results. Disclosures Alexandra Stone, PhD, United Health Group: Stocks/Bonds (Public Company) Wafa Tarazi, PhD, MHPA, UnitedHealth Group: I am an employee at UHG Nwanneamaka Ume, MPH, Optum: I am an Employee of Optum, a UnitedHealth Group subsidiary Andrea Steffens, MPH, OPTUM: Employee|UnitedHealth Group: Stocks/Bonds (Public Company) Katherine Andrade, MPH, Optum: Employee|United Health Group: Stocks/Bonds (Public Company) Garrett Gremel, MS, Optum: Employee Ami R. Buikema, MPH, Optum: Employee|UnitedHealth Group: Stocks/Bonds (Public Company) Wenya Grace Yang, MPA, MA, UnitedHealth Group: I am an employee of UHG.|UnitedHealth Group: Stocks/Bonds (Public Company)

P-1949. The Self-Controlled Risk Interval: An Evolving Method for Vaccine Effectiveness Estimation in Complex Real-World Data

Abstract Background Post-marketing assessments of vaccine effectiveness (VE) are critical to informing health policy, understanding real-world vaccine performance, and ensuring confidence in vaccination efforts, though imperfect comparator groups can limit the accuracy of VE assessments. The self-controlled risk interval (SCRI) design, a case-only analytic method historically used in safety studies, controls for time-invariant confounding using within-person comparisons and has recently been used to assess VE. This targeted review identifies and characterizes SCRI’s growing role in VE research.Table 1:Vaccine Effectiveness Studies that Used the Self-Controlled Risk Interval Design Methods Targeted database searches were conducted in Medline, Embase®, Embase Preprints, and BIOSIS Previews®, and the preprint servers: medRxiv, preprints with The Lancet, and Research Square. Key terms (and related variations) included: vaccine effectiveness, COVID-19 vaccines, and self-controlled risk interval. Results Two VE studies that used the SCRI design were identified (Table 1). Both were retrospective database studies (one using electronic health records and the other insurance claims data), evaluated SARS-CoV-2 infection, and used similar post-vaccination control windows, but featured different post-vaccination risk windows and were conducted among notably different populations. An additional 10 identified studies included post-vaccination person-time in the comparison group but did not make within-person comparisons to address important potential confounding, acknowledging that post-vaccination benefits are not appreciated immediately. These studies varied in the study populations, outcomes assessed, comparison group composition, and analytic methods. Conclusion Use of SCRI for VE estimation has been limited; however, the increasing inclusion of post-vaccination control arm person-time in other COVID-19 VE studies merits the broadening of its application. The results of this review suggest that, in the context of COVID-19, SCRI VE evaluation is a promising tool where there are no suitable comparison groups, increasingly complex individual backgrounds (e.g., infection history, number and timing of doses, or heterologous vaccination), or insufficient information for confounding control. Disclosures Jonathan Fix, PhD, Novavax, Inc.: employee|Novavax, Inc.: Stocks/Bonds (Public Company) Anthony M. Marchese, PhD, Novavax Inc: Employee|Novavax Inc: Stocks/Bonds (Public Company) Hadi Beyhaghi, MD, PhD, Novavax, Inc.: employee|Novavax, Inc.: Stocks/Bonds (Public Company) Matthew D. Rousculp, PhD, Novavax, Inc.: employee|Novavax, Inc.: Stocks/Bonds (Public Company)

P-672. Coinfections with Respiratory Syncytial Virus (RSV) in a Cohort of Adults with Pre-Existing Comorbidities in Wisconsin from 2015-16 through 2019-20

Abstract Background Current evidence suggests that coinfections with respiratory syncytial virus (RSV) may affect symptom and disease severity during acute respiratory infection (ARI). However, this data is largely observed in children, with limited data regarding RSV coinfections in adults with pre-existing comorbidities.Table 1.Demographic and clinical characteristics among high-risk adults with RSV coinfection and RSV single infection from 2015-16 through 2019-20 respiratory virus seasons. Methods We conducted a retrospective cohort study, using existing data and respiratory specimens from community-dwelling adult participants in a test-negative study of influenza vaccine effectiveness. The study was conducted in an outpatient setting in Wisconsin from 2015-2016 through 2019-2020. Participants included in this analysis had ≥1 pre-existing comorbidity (Table 1). Residual respiratory specimens were retested using a PCR-based multiplex panel for adenovirus (HAdV), seasonal coronaviruses (HCoV), human metapneumovirus, rhinovirus/enterovirus (HRV), influenza virus (IFV) (influenza A H1pdm09, H3, and B), human parainfluenza virus types 1-4, RSV A and B. RSV coinfection was defined as the detection of two or more pathogens, one of which was RSV A or B and the other non-RSV. We used chi-square tests to compare demographic and clinical characteristics between those with RSV coinfection vs. RSV single infection. Results Of 3601 respiratory samples from high-risk adults tested, RSV coinfection was detected in 18 (0.5%) participants and RSV single infection in 285 (7.9%). RSV coinfections occurred with HRV (8; 44.4%); IFV (6; 33.3%); HCoV (5; 27.8%); and HAdV (1; 5.6%), and was highest among those aged 18-49 years (9; 50%). We did not observe statistically significant differences between coinfections vs. single infections by presence of upper or lower respiratory symptoms, or by high-risk conditions (Table 1). Hospitalization occurred only in individuals with RSV single infections (10, 3.5%). Conclusion RSV viral co-infections were detected in 0.5% of 3601 respiratory samples of high-risk adults, including coinfections with HRV, IFV, HCoV and HAdV. We did not observe significant differences in symptoms between those with RSV coinfection vs single infection, however, hospitalization occurred only in individuals with RSV single infection. Limitations of this analysis include its sample size and the restrictions of multiplex assay sensitivity and specificity. Disclosures Oluwakemi Alonge, MPH, CPH, CSL Seqirus: Grant/Research Support|GSK Inc.: Grant/Research Support Maria Sundaram, PhD, MSPH, GSK Inc.: Grant/Research Support|Pfizer Inc.: Grant/Research Support Huong Nguyen, PhD, MPH, CSL Seqirus: Advisor/Consultant|CSL Seqirus: Grant/Research Support|GSK: Grant/Research Support|ModernaTX: Advisor/Consultant|ModernaTX: Grant/Research Support Jennifer P. King, MPH, GSK Inc.: Grant/Research Support|ModernaTX: Grant/Research Support Elisha Stefanski, BS, CSL Seqirus: Grant/Research Support|GSK Inc.: Grant/Research Support|ModernaTX: Grant/Research Support Pouya Saeedi, PhD., GSK: Employment Yves Brabant, MEng, GSK: Employed Jean-Yves Pirçon, PhD, GSK Inc.: Employment

P-1189. Clinical Characteristics and Outcomes of Medically Attended Respiratory Syncytial Virus (RSV) Infections in Children 24-59 months of Age

Abstract Background RSV infection is a major cause of lower respiratory tract disease (LRTD) among young children. However, clinical features of RSV disease in children ≥24 months of age are not well described. This study assessed the signs, symptoms, and clinical characteristics of RSV in children 24-59 months of age in both outpatient and inpatient settings. Methods We abstracted clinical records from RSV-positive children 24-59 months of age who sought outpatient care for acute respiratory illness (ARI) and who had provided a respiratory swab for RSV testing by RT-PCR as part of influenza vaccine effectiveness studies from 2004-10 and 2014-15. Records were also abstracted for hospitalized children with laboratory-confirmed RSV from 2010-23. We analyzed clinical features of RSV in each care setting. Results We identified 580 pediatric patients 24-59 months with medically attended, laboratory-confirmed RSV infection, including 419 outpatient and 161 inpatient encounters. The prevalence of comorbidities was 31% among outpatients and 66% among inpatients. The median hospital stay was 3.0 days (IQR: 2.0 - 4.3), and 22% were admitted to the pediatric intensive care unit (PICU). Abnormal lung sounds were observed in 21% of outpatients and 70% of hospitalized patients (P< 0.001). Shortness of breath was documented in 6% of outpatients and in 71% of inpatients (P< 0.001). Increased respiratory rate was recorded in 10% of outpatients and in 48% of inpatients (P< 0.001). The median oxygen saturation at the first measurement for outpatients was 96% (IQR: 95%–97%), compared to 93% (IQR: 90%–95%) for inpatients (P < 0.001). In the complete dataset, 25% of outpatients and 93% of hospitalized children aged 24-59 months exhibited signs or symptoms of LRTD, such as retractions, abnormal lung sounds, hypoxemia and/or need for supplemental oxygen. Conclusion Among children 24-59 months of age with medically attended RSV infection, two-thirds of outpatient visits and one-third of inpatient visits occurred in previously healthy children. Nearly one-quarter of hospitalized children required PICU admission, indicating significant morbidity in this age group. Further research on the epidemiology of RSV in children outside of infancy is needed. Disclosures Zhe Zheng, PhD, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Jennifer P. King, MPH, GSK Inc.: Grant/Research Support|ModernaTX: Grant/Research Support Linwei Li, n/a, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Tianyu Sun, n/a, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Catherine A. Panozzo, PhD, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Thomas G. Boyce, MD, GSK: Grant/Research Support|Moderna, Inc.: Grant/Research Support Iliana Leony Lasso, n/a, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Sabine Schnyder Ghamloush, MD, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Sonia K. Stoszek, PhD, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Matthew D. Snape, MBBS, MD, FRCPCH, FPM, FMedSci, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Evan J. Anderson, MD, Moderna, Inc.: Employee|Moderna, Inc.: Stocks/Bonds (Public Company) Huong McLean, PhD, MPH, CSL Seqirus: Advisor/Consultant|CSL Seqirus: Grant/Research Support|GSK: Grant/Research Support|ModernaTX: Advisor/Consultant|ModernaTX: Grant/Research Support Joshua Petrie, PhD, CSL Seqirus: Grant/Research Support|Moderna, Inc.: Grant/Research Support

P-718. Getting to the Heart of Respiratory Syncytial Virus Disease Burden

Abstract Background There is increasing evidence that viral respiratory infections are associated with acute cardiovascular events. The majority of studies to date involve SARS-CoV-2 and influenza, whereas data for RSV are limited. Methods We performed an analysis of electronic records from patients with documented RSV hospitalization from a prospective incidence study conducted in Rochester, NY from 2017-2020. Clinical, demographic, medical history, and cardiac risk factor data were collected. Outpatient and hospital encounters were reviewed for cardiovascular events (CVE) 6 months before RSV infection and during “high-risk” and “late-risk” periods, 28 days and 1-5 months, respectively, after RSV for a total of one year of observation. Events of interest were myocardial infarction (MI), new or exacerbation of congestive heart failure (CHF), new atrial fibrillation (Afib) or with rapid ventricular response (RVR), arrhythmias, stroke, and others. Peak cardiac biomarkers (troponin and brain natriuretic peptide [BNP]) were collected. Chart review was conducted by 2 physicians independently and consensus achieved. Cardiac Risk Factors Cardiac Risk factors: HTN, hyperlipidemia, diabetes mellitus, family history of CAD in a first-degree relative, current smoking, BMI>30. Results RSV was diagnosed in 471 hospitalized adults. The median age was 69 years, 59% were female and nearly all had underlying medical conditions (92%) with hypertension and diabetes mellitus most common (Figure 1). Cardiac risk factors are summarized in Table 1. During a high-risk period, 167 patients had 270 CVE (89% within 1 week) and 29 patients died. The most common complications were new or exacerbated CHF (27.5%) and Afib with RVR (12.9%), Figure 2. The rates of CVE during pre-RSV, high-risk, and late-risk periods were 0.34, 6.9, and 0.45 per person-years, respectively. Notably, the Incidence Rate Ratio (IRR) of CVE during the high-risk period was 19.9 (CI 15.5-25.9) compared to the pre-RSV period, p< 0.0001. IRR of CVE during late-risk period was 1.55 (CI 1.1-2.1) compared to pre-RSV period, p=0.003. Increased troponins occurred in 34% and high BNP in 62% of those tested. Total number of Cardiovascular Events (CE) was 270 events in 167 patients. CHF – congestive heart failure new onset or decompensation requiring intervention; STEMI/NSTEMI – ST-Elevation Myocardial Infarction, Non-ST Elevation Myocardial Infarction; Afib/Aflutter – Atrial Fibrillation or Atrial Flutter new onset or with rapid ventricular response requiring intervention; SVT – supraventricular tachycardia requiring intervention; VT/NSVT/PEA – Ventricular Tachycardia, Non-sustained Ventricular Tachycardia, Pulseless Electrical Activity; PE/DVT – Pulmonary embolism, Deep Venous Thrombosis; CVA – Cerebrovascular Accident. Conclusion RSV infection is associated with decompensation of chronic cardiac conditions as well as new CVEs with effects persisting past a traditional 28-day risk period. Prevented CVEs should be assessed in upcoming adult RSV vaccine effectiveness studies. Disclosures Edward E. Walsh, MD, Enanta: Advisor/Consultant|Enanta: Honoraria|GSK: Advisor/Consultant|Janssen: Advisor/Consultant|Merck: Advisor/Consultant|Merck: Grant/Research Support|Pfizer: Advisor/Consultant|Pfizer: Grant/Research Support Angela R. Branche, MD, Cyanvac: Grant/Research Support|GSK: Advisor/Consultant|Merck: Grant/Research Support|Moderna: Grant/Research Support|Moderna: Honoraria|Novavax: Advisor/Consultant|Pfizer: Grant/Research Support|Sanofi: Honoraria|VaxCo: Grant/Research Support Ann R. Falsey, MD, ADMA: Board Member|BioFire Diagnostics: Grant/Research Support|CyanVac: Grant/Research Support|GSK: Advisor/Consultant|GSK: Travel support|Janssen: Grant/Research Support|Moderna: Advisor/Consultant|Moderna: Grant/Research Support|Pfizer: Advisor/Consultant|Pfizer: Grant/Research Support|Sanofi Pasteur: Advisor/Consultant|Sanofi Pasteur: Travel support|VaxCo: Grant/Research Support

Methods to Adjust for Confounding in Test-Negative Design COVID-19 Effectiveness Studies: Simulation Study.

Real-world COVID-19 vaccine effectiveness (VE) studies are investigating exposures of increasing complexity accounting for time since vaccination. These studies require methods that adjust for the confounding that arises when morbidities and demographics are associated with vaccination and the risk of outcome events. Methods based on propensity scores (PS) are well-suited to this when the exposure is dichotomous, but present challenges when the exposure is multinomial. This simulation study aimed to investigate alternative methods to adjust for confounding in VE studies that have a test-negative design. Adjustment for a disease risk score (DRS) is compared with multivariable logistic regression. Both stratification on the DRS and direct covariate adjustment of the DRS are examined. Multivariable logistic regression with all the covariates and with a limited subset of key covariates is considered. The performance of VE estimators is evaluated across a multinomial vaccination exposure in simulated datasets. Bias in VE estimates from multivariable models ranged from -5.3% to 6.1% across 4 levels of vaccination. Standard errors of VE estimates were unbiased, and 95% coverage probabilities were attained in most scenarios. The lowest coverage in the multivariable scenarios was 93.7% (95% CI 92.2%-95.2%) and occurred in the multivariable model with key covariates, while the highest coverage in the multivariable scenarios was 95.3% (95% CI 94.0%-96.6%) and occurred in the multivariable model with all covariates. Bias in VE estimates from DRS-adjusted models was low, ranging from -2.2% to 4.2%. However, the DRS-adjusted models underestimated the standard errors of VE estimates, with coverage sometimes below the 95% level. The lowest coverage in the DRS scenarios was 87.8% (95% CI 85.8%-89.8%) and occurred in the direct adjustment for the DRS model. The highest coverage in the DRS scenarios was 94.8% (95% CI 93.4%-96.2%) and occurred in the model that stratified on DRS. Although variation in the performance of VE estimates occurred across modeling strategies, variation in performance was also present across exposure groups. Overall, models using a DRS to adjust for confounding performed adequately but not as well as the multivariable models that adjusted for covariates individually.

Estimation of the adjusted risk difference for very rare events, large samples, and extreme exposure frequency: Application of Vaccine Effectiveness, Networking, and Universal Safety study data

Estimation of the adjusted risk difference for very rare events, large samples, and extreme exposure frequency: Application of Vaccine Effectiveness, Networking, and Universal Safety study data

Relative effectiveness of the second booster COVID-19 vaccines against laboratory confirmed SARS-CoV-2 infection in healthcare workers: VEBIS HCW VE cohort study (1 October 2022-2 May 2023).

Repeated COVID-19 booster vaccination was recommended in healthcare workers (HCWs) to maintain protection. We measured the relative vaccine effectiveness (rVE) of the second booster dose of COVID-19 vaccine compared to the first booster, against laboratory-confirmed SARS-CoV-2 infection in HCWs. In a prospective cohort study among HCWs from 12 European hospitals, we collected nasopharyngeal or saliva samples at enrolment and during weekly/fortnightly follow-up between October 2022 and May 2023. We estimated rVE of the second versus first COVID-19 vaccine booster dose against SARS-CoV-2 infection, overall, by time since second booster and restricted to the bivalent vaccines only. Using Cox regression, we calculated the rVE as (1-hazard ratio)*100, adjusting for hospital, age, sex, prior SARS-CoV-2 infection and at least one underlying condition. Among the 979 included HCWs eligible for a second booster vaccination, 392 (40%) received it and 192 (20%) presented an infection during the study period. The rVE of the second versus first booster dose was -5% (95%CI: -46; 25) overall, 3% (-46; 36) in the 7-89days after receiving the second booster dose. The rVE was 11% (-43; 45) when restricted to the use of bivalent vaccines only. The bivalent COVID-19 could have reduced the risk of SARS-CoV-2 infection among HCWs by 11%. However, we note the limitation of imprecise rVE estimates due to the proportion of monovalent vaccine used in the study, the small sample size and the study being conducted during the predominant circulation of XBB.1.5 sub-lineage. COVID-19 vaccine effectiveness studies in HCWs can provide important evidence to inform the optimal timing and the use of updated COVID-19 vaccines.

Magnitude of Potential Biases in COVID-19 Vaccine Effectiveness Studies due to Differential Healthcare seeking following Home Testing: Implications for Test Negative Design Studies.

The test-negative design (TND) is widely used to estimate COVID-19 vaccine effectiveness (VE). Biased estimates of VE may result from effects of at-home SARS-CoV-2 rapid diagnostic test (RDT) results on decisions to seek healthcare. To investigate magnitude of potential bias, we constructed decision trees with input probabilities obtained from longitudinal surveys of U.S. adults between March 2022 - October 2023. Prevalence of at-home RDT use and healthcare seeking following a positive or negative RDT result was estimated by participant vaccination status and socio-demographic characteristics. At true VE values ranging from 5% to 95%, we defined bias as the difference between the observed and true VE . Among 1,918 symptomatic adults, prevalence of at-home RDT use was higher among vaccinated (37%) versus unvaccinated (22%) participants. At-home RDT use was associated with seeking care, and participants reporting positive RDT were more likely than those reporting negative RDT to have sought care when ill. In primary analyses, we observed downward bias in VE estimates that increased in magnitude when true VE was low. Variations in proportions of vaccination, at-home RDT use and healthcare seeking by socio-demographic characteristics may impact VE estimates. Further evaluation of potential impact of at-home RDT use on VE estimates is warranted.

Factors associated with COVID-19 in-hospital death and COVID-19 vaccine effectiveness against COVID-19 hospitalization in the Philippines during pre-omicron and omicron period: A case-control study (MOTIVATE-P study).

COVID-19 vaccine effectiveness (VE) studies are limited in low- and middle-income countries. A case-control study was conducted among COVID-19 and other pneumonia patients admitted to a hospital in the Philippines during the pre-Omicron and Omicron periods. To elucidate factors associated with in-hospital death, 1782 COVID-19 patients were assessed. To estimate absolute VE for various severe outcomes, 1059 patients were assessed (869 [82.1%] COVID-19 cases; 190 [17.9%] controls). Factors associated with in-hospital death included older age, tuberculosis (adjusted odds ratio [aOR] 2.45 [95% confidence interval {95% CI} 1.69-3.57]), HIV (aOR 3.30 [95% CI 2.03-5.37]), and current smokers (aOR 2.65 [95% CI 1.72-4.10]). Pre-Omicron, the primary series provided high protection within a median of 2 months (hospitalization: 85.4% [95% CI 35.9-96.7%]; oxygen requirement: 91.0% [95% CI 49.4-98.4%]; invasive mechanical ventilation (IMV): 97.0% [95% CI 65.7-99.7%]; death: 96.5% [95% CI 67.1-99.6%]). During Omicron, the primary series provided moderate-high protection within a median of 6-9 months (hospitalization: 70.2% [95% CI 27.0-87.8%]; oxygen requirement: 71.4% [95% CI 29.3-88.4%]; IMV: 72.7% [95% CI -11.6-93.3%]; death: 58.9% [95% CI -82.8-90.8%]). Primary series VE against severe COVID-19 outcomes was consistently high for both pre-Omicron and Omicron in a setting where approximately half of the vaccinees received inactivated vaccines.

Bias in control selection associated with the use of rapid tests in influenza vaccine effectiveness studies.

In test-negative design studies that use rapid tests to estimate influenza vaccine effectiveness (VE) a common concern is case/control misclassification due to imperfect test sensitivity and specificity. However, an imperfect test can also fail to exclude from the control group people that do not represent the source population, including people infected with other influenza types or other vaccine-preventable respiratory viruses for which vaccination status is correlated. We investigated these biases by comparing the effectiveness of seasonal 2023/24 influenza vaccination against influenza A and B based on PCR versus rapid test results, excluding controls who tested positive for SARS-CoV-2 or the other type of influenza. By PCR, VE against influenza A was 49% (95%CI 26-65%) after exclusion of PCR-confirmed influenza B and SARS-CoV-2 controls. Corresponding VE against influenza B was 65% (95%CI 35-81%). VE estimated by adjusting for COVID-19 vaccination status yielded similar estimates to the scenario that excluded SARS-CoV-2-positive controls. When case/control status and exclusions from test-negative controls were determined by rapid test, VE was reduced by 5-15 percentage points. Bias correction methods were able to reduce these discrepancies. When estimating VE from a test-negative study using rapid test results, methods to correct misclassification bias are recommended.

Community intervention of a single-dose or 2-dose regimen of bivalent human papillomavirus vaccine in schoolgirls in Thailand: vaccine effectiveness 2 years and 4 years after vaccination.

With accumulating evidence of single-dose human papillomavirus (HPV) vaccine efficacy in young women, we conducted a community vaccine effectiveness study comparing HPV single-dose and 2-dose regimens (0 and 6 months) of a bivalent HPV vaccine among grade 8 schoolgirls (aged 13-14 years) in Thailand. In 2018, eligible grade 8 schoolgirls in Udon Thani (single dose) and Buri Ram (2 doses) provinces were offered HPV vaccine per assigned dose regimen. Concurrently, a cross-sectional survey for measuring baseline HPV prevalence was conducted in grade 10 (n = 2600) and grade 12 unvaccinated schoolgirls (n = 2000) in each province. HPV infection was assessed in first-void urine samples, tested by DNA polymerase chain reaction on the cobas 4800 system (Roche Molecular Diagnostics, Pleasanton, CA). All samples positive on the cobas system and an equal number of negative samples were also tested by Anyplex II HPV28 Detection (Seegene, Seoul, South Korea). The surveys were repeated in 2020 and 2022, when vaccinated grade 8 schoolgirls reached grade 10, and then subsequently grade 12, respectively. Vaccine effectiveness was estimated by comparing the weighted prevalence of HPV-16 or HPV-18 between grade-matched unvaccinated schoolgirls on the baseline survey (2018) and vaccinated schoolgirls in the year-2 (2020) and year-4 (2022) surveys. Adjustment methods were used in the analysis to account for potential differences in sexual behavior due to the noncontemporaneous comparison. The prevalence of HPV-16 and HPV-18 on the baseline survey among unvaccinated grade 10/grade 12 schoolgirls was 2.90% (95% confidence interval [CI] = 2.54% to 3.31%)/3.98% (95% CI = 3.52% to 4.49%) for Udon Thani and 3.87% (95% CI = 3.46% to 4.34%)/6.13% (95% CI = 5.56% to 6.75%) for Buri Ram. On the year-2 survey, the prevalence among vaccinated grade 10 schoolgirls was 0.57% (95% CI = 0.42% to 0.77%) for Udon Thani and 0.31% (95% CI = 0.21% to 0.47%) for Buri Ram. The 2-year postvaccination crude vaccine effectiveness for the single-dose regimen was estimated at 80.4% (95% CI = 73.9% to 86.9%), and for the 2-dose regimen at 91.9% (95% CI = 88.5% to 95.4%). On the year-4 survey, the prevalence among vaccinated grade 12 schoolgirls was 0.37% (95% CI = 0.25% to 0.56%) for Udon Thani and 0.28% (95% CI = 0.18% to 0.45%) for Buri Ram. Four-year postvaccination crude vaccine effectiveness for the single-dose regimen was estimated at 90.6% (95% CI = 86.6% to 94.6%) and for the 2-dose regimen was estimated at 95.4% (95% CI = 93.2% to 97.6%). All adjustment methods minimally affected vaccine effectiveness for the single-dose and 2-dose regimens. At 4 years after vaccination, the difference in crude vaccine effectiveness between the single-dose and 2-dose regimens was ‒4.79% (95% CI = ‒9.32% to ‒0.25%), meeting the study's noninferiority criteria. Our study demonstrated that both single-dose and 2-dose HPV vaccination significantly decreased HPV-16/18 point prevalence 2 years and 4 years after vaccination. Crude vaccine effectiveness at 4 years after vaccination was greater than 90% for both the single-dose and 2-dose regimens; the single-dose regimen was not inferior to the 2-dose regimen. These data show that a single dose of HPV vaccine provides high levels of protection when administered to schoolgirls younger than 15 years of age.

Biases in COVID-19 vaccine effectiveness studies using cohort design.

Observational studies on COVID-19 vaccine effectiveness (VE) have provided critical real-world data, informing public health policy globally. These studies, primarily using pre-existing data sources, have been indispensable in assessing VE across diverse populations and developing sustainable vaccination strategies. Cohort design is frequently employed in VE research. The rapid implementation of vaccination campaigns during the COVID-19 pandemic introduced differential vaccination influenced by sociodemographic disparities, public policies, perceived risks, health-promoting behaviors, and health status, potentially resulting in biases such as healthy user bias, healthy vaccinee effect, frailty bias, differential depletion of susceptibility bias, and confounding by indication. The overwhelming burden on healthcare systems has escalated the risk of data inaccuracies, leading to outcome misclassifications. Additionally, the extensive array of diagnostic tests used during the pandemic has also contributed to misclassification biases. The urgency to publish quickly may have further influenced these biases or led to their oversight, affecting the validity of the findings. These biases in studies vary considerably depending on the setting, data sources, and analytical methods and are likely more pronounced in low- and middle-income country (LMIC) settings due to inadequate data infrastructure. Addressing and mitigating these biases is essential for accurate VE estimates, guiding public health strategies, and sustaining public trust in vaccination programs. Transparent communication about these biases and rigorous improvement in the design of future observational studies are essential.

Improved Methods for Vaccine Effectiveness Studies.

Evaluating the impact of public health investments in vaccination programs is crucial for ensuring their efficiency and effectiveness. Vaccine effectiveness (VE) studies, such as those using the test-negative design (TND), are commonly used to confirm the impact of vaccines and to guide future improvements. The TND, favored for its simplicity and cost efficiency, mitigates biases common in other epidemiological study designs. However, its validity can be compromised by inconsistent symptom definitions and retrospective data application. Here we summarize recent findings that highlight the need to address correlated vaccination behaviors when estimating VE, that suggest using negative control variables to reduce confounding, and that recommend accounting for prior infection history in VE studies to improve accuracy and reliability. These insights are important for refining VE estimation methods.

Quantifying and Adjusting for Confounding From Health-Seeking Behavior and Health Care Access in Observational Research.

Health-seeking behavior and health care access (HSB/HCA) are recognized confounders in many observational studies but are not directly measurable in electronic health records. We used proxy markers of HSB/HCA to quantify and adjust for confounding in observational studies of influenza and COVID-19 vaccine effectiveness (VE). This cohort study used primary care data prelinked to secondary care and death data in England. We included individuals aged ≥66 years on 1 September 2019 and assessed influenza VE in the 2019-2020 season and early COVID-19 VE (December 2020-March 2021). VE was estimated with sequential adjustment for demographics, comorbidities, and 14 markers of HSB/HCA. Influenza vaccination in the 2019-2020 season was also considered a negative control exposure against COVID-19 before COVID-19 vaccine rollout. We included 1 991 284, 1 796 667, and 1 946 943 individuals in the influenza, COVID-19, and negative control exposure populations, respectively. Markers of HSB/HCA were positively correlated with influenza and COVID-19 vaccine uptake. For influenza, adjusting for HSB/HCA markers in addition to demographics and comorbidities increased VE against influenza-like illness from -1.5% (95% CI, -3.2% to .1%) to 7.1% (95% CI, 5.4%-8.7%) with a less apparent trend for more severe outcomes. For COVID-19, adjusting for HSB/HCA markers did not change VE estimates against infection or severe disease (eg, 2 doses of BNT162b2 against infection: 82.8% [95% CI, 78.4%-86.3%] to 83.1% [95% CI, 78.7%-86.5%]). Adjusting for HSB/HCA markers removed bias in the negative control exposure analysis (-7.5% [95% CI, -10.6% to -4.5%] vs -2.1% [95% CI, -6.0% to 1.7%] before vs after adjusting for HSB/HCA markers). Markers of HSB/HCA can be used to quantify and account for confounding in observational vaccine studies.

Symptoms Six Weeks After COVID-19 Are Reduced Among US Health Care Personnel Receiving Additional Vaccine Doses During the Omicron Period, December 2021-April 2022.

The objective of this study was to test the hypothesis that subsequent doses of the coronavirus disease 2019 (COVID-19) vaccine are associated with lower incidence of COVID-19-like symptoms at 6 weeks after infection. This study was a case-control analysis of health care personnel in an ongoing multicenter COVID-19 vaccine effectiveness study. We enrolled participants at the time of COVID-19-like symptoms between December 19, 2021, and April 27, 2022, which corresponded to the early Omicron-predominant period after original monovalent severe acute respiratory syndrome coronavirus 2 additional vaccination doses became available. Our outcome was self-reported symptoms completed 6 weeks after the onset of symptoms. We enrolled 2478 participants, of whom 1422 (57%) had COVID-19. The prevalence of symptoms at 6 weeks was 26% (n = 373) in those with COVID-19 and 18% (n = 195) in those without COVID-19. Fatigue (11%) and difficulty sleeping (7%) were most strongly associated with COVID-19. A total of 1643 (66%) participants received a subsequent vaccine dose (after the primary series). Participants with COVID-19 who had received a subsequent vaccination had lower odds of symptoms at 6 weeks (adjusted odds ratio [aOR], 0.55; 95% CI, 0.43-0.70), but this relationship was not observed in those without COVID-19 (aOR, 0.87; 95% CI, 0.59-1.29). Health care personnel who received subsequent doses of original monovalent COVID-19 vaccine had a lower prevalence of symptoms at 6 weeks than those that did not.

Safety of quadrivalent recombinant influenza vaccine in pregnant persons and their infants

BackgroundThe Advisory Committee on Immunization Practices (ACIP) and the American College of Obstetricians and Gynecologists (ACOG) recommend that all pregnant persons receive any licensed, recommended, and age-appropriate inactivated influenza vaccine (SD-IIV) or recombinant influenza vaccine (RIV) to protect against influenza and influenza-related complications. RIV was safe and efficacious in pre- and postlicensure studies, however there is limited RIV safety data in pregnant persons. ObjectiveTo evaluate the safety of quadrivalent recombinant influenza vaccine (RIV4) versus a quadrivalent standard-dose, inactivated influenza (SD-IIV4) in a large cohort of pregnant persons and their infants. Study DesignThis postlicensure observational safety study conducted at Kaiser Permanente Northern California evaluated the subset of pregnant persons vaccinated in routine care as part of a larger cluster-randomized vaccine effectiveness study comparing RIV4 vs. SD-IIV4 (ClinicalTrials.gov NCT03694392). We identified pregnancy (spontaneous abortion, preterm labor, stillbirth/fetal death, congenital/fetal anomalies detected during pregnancy, eclampsia/pre-eclampsia, placental abruption), birth (preterm birth, low birth weight, small for gestational age), and neonatal/infant outcomes (infant death, failure to thrive, congenital anomalies detected after delivery) using diagnostic codes among pregnant persons ≥18 years immunized with RIV4 or SD-IIV4 during the 2018/19 and 2019/20 influenza seasons and their infants. We used conditional logistic regression adjusted for age group, race, ethnicity, trimester of influenza vaccination, comorbidities, and BMI, stratified by gestational age to estimate the odds ratio (OR) of pregnancy outcomes following vaccination with RIV4 vs. SD-IIV4. Using logistic regression, we separately estimated the adjusted OR of birth and neonatal/infant outcomes in the first year of life (eg, death) in infants of RIV4 vs. SD-IIV4 vaccinated pregnant persons. ResultsThe study population included 48,781 pregnant persons (RIV4 = 14,981; SD-IIV4 = 33,800) and 47,394 live births (RIV4 = 14,538; SD-IIV4 = 32,856). There was no statistical difference in any pregnancy outcome or in birth and neonatal/infant outcome between RIV4 vs. SD-IIV4 vaccinated pregnant persons and their infants. ConclusionCompared with receipt of a SD-IIV4 during pregnancy, this large study did not identify any pregnancy, birth, or neonatal/infant safety concerns following receipt of a RIV4 during pregnancy and demonstrates that the safety of RIV4 in pregnancy was similar to SD-IIV4. This study provides additional evidence regarding the safety of influenza vaccination in pregnant persons and further supports ACIP and ACOG recommendations that all pregnant persons receive an inactivated or recombinant influenza vaccine.

A real-world study of BBIBP-CorV vaccine effectiveness in a Sri Lanka rural province

ObjectiveReal-world studies assessing the effectiveness of the BBIBP-CorV vaccine in low and middle-income countries are limited. We evaluated the BBIBP-CorV vaccine's effectiveness in reducing COVID-19 symptomatic disease, hospitalisation, severe disease, and mortality during the third wave of the pandemic in Sri Lanka. MethodsWe conducted a test-negative case-control study in North Central Province from May 2021 to February 2022. Evidence of vaccination was obtained from the national registry. The PCR-positive patients were cases, while negative individuals were controls. Adjusted vaccine effectiveness (aVE) was computed for fully, partially, and non-vaccinated groups in reducing symptomatic disease, hospitalisation, severe disease, and mortality. ResultsOur study involved 3305 cases and 3418 controls. The overall aVE for preventing PCR-positive infection in fully vaccinated was 30·8 % (95 % CI:17·9–41·6). In fully vaccinated over 60 years, the overall aVE was 72·3 % (95 % CI: 49·7–84·8). Full vaccination with BBIBP-CorV is effective in reducing hospitalisation, severe COVID-19 disease, and death, with aVE rates of 70·3 % (95 % CI: 57·9–79·0), 88·9 % (95 % CI: 81·8–93·2), and 92·3 % (95 % CI: 84·8–96·1) respectively. ConclusionIndividuals who have received two doses of the BBIBP-CorV vaccine are protected against hospitalisation, severe COVID-19 disease, and death. Duration of protection against hospitalisation, severe COVID-19, and fatal COVID-19 sustained at least 121 days, with no sign of waning during that time.

Pertussis vaccine effectiveness following country-wide implementation of a hexavalent acellular pertussis immunization schedule in infants and children in Panama

ABSTRACT Despite high pediatric vaccination coverage rates (VCRs), pertussis incidence has increased worldwide, including in several countries in Latin America in the last two decades. Given the few vaccine effectiveness (VE) studies in Latin American countries, this retrospective, observational, cohort study estimated the effectiveness of hexavalent acellular (aP) primary and booster vaccination (wP) against pertussis in infants (6.5–18.5 months) and children (18.5–48.5 and 48.5–72.5 months) in Panama. Age-specific incidence rates (IRs) were calculated for the vaccine’s pre-initiation (2001–2013), initiation (2014), and post-initiation (2015–2019) periods. VCRs and trends were determined, and VE was analyzed using a case coverage or screening method to compare proportions of vaccinated cases and vaccinated individuals in the population. Between 2001–2019, 868 confirmed pertussis cases were reported in Panama; 712 (82.0%; 54.8 cases/year) during the pre-initiation period, 19 (2.2%; 19 cases/year) during the initiation period, and 137 (15.8%; 27.4 cases/year) during the post-initiation period. Panama underwent cyclical increases in IRs, which varied between age groups. VCRs increased for primary and booster doses. Between 2015 and 2019, third-dose yearly vaccine coverage increased, on average, 3.3%. Specifically, during the post-initiation period, 109/137 (79.6%) of cases were unvaccinated. Relative VE was estimated at 96.2% [95% CI: 86.5%, 98.9%] with three doses; 100% with 4 and 5 booster doses. Absolute VE was estimated at 99.3% with three doses only. These results show that vaccination played an important role in maintaining a low number of pertussis cases in Panama, affirming the need for sustained investment and commitment to vaccination programs.

Potential Biases in Test-Negative Design Studies of COVID-19 Vaccine Effectiveness Arising from the Inclusion of Asymptomatic Individuals.

The test-negative design (TND) is a popular method for evaluating vaccine effectiveness (VE). A "classical" TND study includes symptomatic individuals tested for the disease targeted by the vaccine to estimate VE against symptomatic infection. However, recent applications of the TND have attempted to estimate VE against infection by including all tested individuals, regardless of their symptoms. In this article, we use directed acyclic graphs and simulations to investigate potential biases in TND studies of COVID-19 VE arising from the use of this "alternative" approach, particularly when applied during periods of widespread testing. We show that the inclusion of asymptomatic individuals can potentially lead to collider stratification bias, uncontrolled confounding by health and healthcare-seeking behaviors (HSBs), and differential outcome misclassification. While our focus is on the COVID-19 setting, the issues discussed here may also be relevant in the context of other infectious diseases. This may be particularly true in scenarios where there is either a high baseline prevalence of infection, a strong correlation between HSBs and vaccination, different testing practices for vaccinated and unvaccinated individuals, or settings where both the vaccine under study attenuates symptoms of infection and diagnostic accuracy is modified by the presence of symptoms.