Background: The CDC recommends multiple doses of COVID-19 vaccine for children aged 6 months–5 years (under 5). However, limited information exists on hesitancy toward the original vaccine or boosters since the updated vaccines became available for this age group. This study fills this gap by utilizing the national-level Household Pulse Survey conducted from March 1 to May 8, 2023 (Phase 3.8).

Methods: The sampled housing units were extracted from the Census Bureau’s Master Address File. To calculate proportions and survey-weighted 95% confidence intervals, 80 replicate weights derived from successive difference replications were used. The study investigated the percentage of children under 5 who had received at least one dose of the COVID-19 vaccine, evaluating the probability of vaccination categorized by willingness to vaccinate (definitely or probably), moderate hesitancy (unsure or probably not), and strong hesitancy (definitely not), while also examining sociodemographic factors influencing vaccination decision-making.

Results: Overall, 24.5% (95% CI: 22.2–26.8) of children under 5 received at least one COVID-19 vaccine dose, with the highest percentage among non-Hispanic Asians (42.1%, 95% CI: 36.4–47.8). Vaccine hesitancy (moderate hesitancy, strong hesitancy) was significantly associated with decreased inclination to vaccinate among household respondents. Decreases in household income and educational attainment correlated with increased strong vaccine hesitancy. Health insurance status was linked to higher vaccine uptake and lower strong hesitancy compared to households without insurance.

Conclusions: Healthcare stakeholders can play vital roles in informing, influencing, and administering COVID-19 vaccines to children under 5, necessitating targeted interventions to combat vaccine hesitancy and emphasize booster dose.

Event history modeling, such as survival and Cox proportional hazard models, is a valuable tool for assessing impact on the timing of a health-related event. This methodology is underutilized in public health campaign evaluation and holds promise for evaluating campaign impact on behavior change. This presentation will discuss the use of event history models to assess campaign impact on the timing of health behavior change using the We Can Do This COVID-19 campaign promotion of monovalent booster vaccination as a case study.

In a longitudinal survey, respondents reported their dates of COVID-19 vaccination and booster uptake from 2021 to 2022. These data were combined with campaign media data, a proxy for campaign exposure, to assess campaign impact on monovalent COVID-19 booster uptake using a Cox proportional hazard model. Analysis found the campaign influenced both the likelihood and timing of booster uptake and that the campaign had the most influence among those who had initially delayed getting their primary COVID-19 vaccine.

In the presentation, we also highlight how using prior event timing as a modifying variable in event history models helps explain the impact of public health campaigns on health behaviors. The timing of past behaviors can offer contextual information that enhances the accuracy of campaign evaluation models. This presentation will highlight future directions for the use of event history models in campaign evaluation and how this methodology can be used to examine how campaigns influence the timing of episodic or habitual health behaviors.

For persons living with HIV (PLWH), there are currently no studies that comprehensively consider how the confluence of risk factors, including comorbidities, clinical/ laboratory data, and SDOH, affects the additional risk faced by PLWH in the COVID-19 pandemic. We investigated the characteristics of PLWH that tend to have severe COVID-19 based on data from the National COVID Cohort Collaborative (N3C) study, the largest cohort of individuals with and without COVID-19 in the US.

We used data released from January 1, 2020 to November 2, 2023. Data were divided into training and testing sets. On the training set, we used machine learning methods to select features predicting severe COVID-19 based on demographics, lab, and comorbidities data. After dimension reduction, we applied consensus clustering with k-means to discover subgroups of patients who share similar characteristics. We performed validation using random forests. We reproduced the clusters on the testing set.

We identified, validated, reproduced, and characterized three distinct subgroups of COVID-19 patients with HIV: cluster 1, the mild COVID-19 group, had lower proportions of severe COVID-19; cluster 2, the SDOH group, where characterized by lowest mean for most of the SDOH variables; cluster 3, the severe COVID-19 group, had higher proportions of severe COVID-19.

Individuals who live with HIV that have severe COVID-19 can be sub-classified into groups with differing COVID-19 severity, and SDOH variables.The identified clusters confirm findings of previous clustering studies. The result shows SDOH variables were not only associated with severe COVID-19 outcome, but also varied by race and ethnicity.

Objective: Assess mRNA vaccine effectiveness against SARS-CoV-2 variants across demographic subgroups

Design: Retrospective Matched Cohort Study.

Participants: 360,331 patients seeking SARS-CoV-2-related care between 07/31/2021-10/31/2021 for Delta and 12/15/2021-03/15/2022 for Omicron. The analytic sample was constructed via propensity score matching on vaccination status (unvaccinated, fully vaccinated, boosted). The analytic samples consist of 161,460 (delta) and 103,193 (omicron).

Results: During delta, fully vaccinated individuals had strong protection against hospitalization (78.4%, 95% CI: 73.5-82.4%), ICU admission (76.3%, 95% CI: 64.8-84.1%), and death (79.7%, 95% CI: 67.3-87.4%) compared to unvaccinated individuals (P<.001). During omicron, fully vaccinated had moderate protection (hospitalization:49.8%, 95% CI: 37.4-59.8%, P<.001, ICU: 49.2%, 95% 16.2-69.3%; P= 0.008, death: 31.1%, 95% CI: -15.7-59.0%; P =0.158) and boosted individuals had strong protection (hospitalization: 84.7%, 95% CI: 76.0-90.2%; P<.001, ICU: 88.2%, 95% CI: 62.6-96.3; P<.001, death: 73.0%, 95% CI: 32.6-89.2%;P= 0.005). Across both delta and omicron variants, individuals aged 18-44 consistently had higher protection against adverse outcomes than those aged 65 and over. Similarly, females demonstrated some increased protection compared to males. Fully vaccinated Black individuals experienced a greater decline in protection against omicron compared to White individuals but similar levels of protection after the booster dose.

Conclusions: Findings underscore vaccine effectiveness in reducing severe COVID-19 outcomes and the importance of booster doses in sustaining protection, particularly among certain demographic groups.