Understanding how cells interact to coordinate tissue function and respond to disease remains a central challenge in biology. Recently developed multiplex bioimaging technologies (MxIF) can quantify multiple protein markers at a single-cell resolution in spatially preserved tissue, offering the opportunity to understand how cellular organization relates to disease mechanisms and treatment response. Spatial generalized linear mixed models for cell co-localization offer a flexible framework for these analyses, accommodating cell, image, or subject-level predictors while accounting for correlation arising from repeated cells per image, repeated images per subject, and spatial dependence among nearby cells. However, as MxIF data typically has hundreds of thousands to millions of cells per study, fitting the full spatial model is computationally infeasible due to the large, dense, spatial covariance matrix. We present a scalable inference strategy that uses parallelizable eigen-decomposition of an empirically estimated spatial correlation matrix to approximate the full spatial model. This approach enables fast estimation of fixed effects of interest without requiring formal estimation of spatial parameters or surface prediction. In simulations, our proposed method could be fitted to large datasets (~285,000 observations) in under a minute, with runtime linearly increasing by only ~7 seconds per 50,000 additional observations. This scalability makes cell-level modeling in the presence of spatial correlation feasible for real-world MxIF datasets, as we demonstrate through applications to metastatic melanoma and high grade serous ovarian carcinoma.

Epidemic models are invaluable tools to understand and implement strategies to control the spread of infectious diseases, as well as to inform public health policies and resource allocation. However, current modeling approaches have limitations that reduce their practical utility, such as the exclusion of human behavioral change in response to the epidemic or ignoring the presence of undetected infectious individuals in the population. These limitations became particularly evident during the COVID-19 pandemic, underscoring the need for more accurate and informative models. To address these challenges, we develop a novel Bayesian epidemic modeling framework to better capture the complexities of disease spread by incorporating behavioral responses and undetected infections. In particular, our framework makes three contributions: (1) leveraging additional data on hospitalizations and deaths in modeling the disease dynamics, (2) accounting for data uncertainty arising from the large presence of asymptomatic and undetected infections, and (3) allowing the population behavioral change to be dynamically influenced by multiple data sources (cases and deaths). We thoroughly investigate the properties of the proposed model via simulation, and illustrate its utility on COVID-19 data from Montréal and Miami.

Background/Objectives

Despite availability of safe and effective COVID-19 vaccines for children, uptake of this vaccine remains low in the United States. Caregiver attitudes and behaviors play a key role in child vaccine decisions, yet little is known about the reasons for vaccine refusal within families or the relationship between COVID-19 vaccine and routine childhood immunization refusal. This study examined 1) associations between caregiver and child COVID-19 vaccination status, 2) alignment of reasons for vaccine non-acceptance between the caregiver themselves and for their child, and 3) associations between COVID-19 vaccine non-acceptance and refusal or delay of routine childhood vaccinations.

Methods

We analyzed data from 6395 families in the NIH ECHO Cohort with self-reported caregiver and child COVID-19 vaccination status between December 2021 and August 2023. Generalized estimating equations were used to assess associations while accounting for cohort clustering.

Results

Of the 6395 families, 25% of caregivers had not accepted the COVID-19 vaccine for themselves, and 59% had not accepted the vaccine for any of their children. Children were more likely to be unvaccinated when their caregivers were unvaccinated (OR: 13.2, 95% CI: 10.3–16.9). Among unvaccinated caregiver-child pairs, reasons for vaccine refusal showed strong concordance. The predominant concern among vaccinated caregivers with unvaccinated children was a belief that the COVID-19 vaccine had not been well tested in children (OR: 3.7, 95% CI: 2.3–6.0). Caregivers who refused the COVID-19 vaccination for themselves were more likely to have delayed or refused routine vaccines for their child (child unvaccinated for COVID-19: OR: 1.8, 95% CI: 1.4–2.1); caregiver unvaccinated for COVID-19: OR 2.8, 95% CI: 2.2–3.4).

Conclusions

COVID-19 vaccine status among caregivers was strongly associated with child COVID-19 vaccine uptake and broader vaccine behaviors. Addressing shared safety concerns and building caregiver trust in vaccines is essential for improving pediatric vaccination coverage.

Cattle and other domestic ruminants are the primary reservoirs of O157 and non-O157 Shiga toxin-producing Escherichia coli (STEC). Living in areas with high ruminant density has been associated with excess risk of infection, which could be due to both direct ruminant contact and residual environmental risk, but the role of each is unclear. We investigated whether there is any meaningful risk to individuals living in ruminant-dense areas if they do not have direct contact with ruminants. Using a Bayesian spatial framework, we investigated the association between the density of ruminants on feedlots and STEC incidence in Minnesota from 2010 to 2019, stratified by serogroup and season, and adjusting for direct ruminant contact. For every additional head of cattle or sheep per 10 acres, the incidence of O157 STEC infection increased by 30% (incidence rate ratio [IRR] 1.30; 95% credible interval [CrI] 1.18, 1.42) or 135% (IRR 2.35; 95% CrI 1.14, 4.20), respectively, during the summer months. Sheep density was also associated with O157 STEC risk during winter (IRR 4.28; 95% CrI 1.40, 8.92). The risk of non-O157 STEC infection was only elevated in areas with goat operations during summer (IRR 19.6; 95% CrI 1.69, 78.8). STEC risk associated with ruminant density was independent of direct ruminant contact across serogroups and seasons. Our findings demonstrate that living in a ruminant-dense area increases an individual’s risk of O157 and non-O157 STEC infection even without direct ruminant contact, indicating that prevention efforts need to extend to community strategies for averting indirect transmission from local ruminant populations.