# Estimating the impact of patient-level risk factors and time-varying hospital unit on healthcare-associated Clostridioides difficile infection using cross-classified multilevel models

**Authors:** Jessica Lynn Webster, Claudine T. Jurkovitz, Brisa N. Sánchez, Stephen Eppes, Neal D. Goldstein

PMC · DOI: 10.1017/ice.2025.10356 · Infection Control and Hospital Epidemiology · 2025-12-01

## TL;DR

This study finds that antibiotic use and hospital unit assignment are key factors in healthcare-associated C. difficile infection.

## Contribution

The novel use of cross-classified multilevel models accounts for patient movement across hospital units.

## Key findings

- Antibiotic exposure is strongly associated with increased odds of C. difficile infection.
- Environmental factors related to hospital units also contribute to infection risk.
- Multilevel modeling reveals the importance of both patient and environmental factors.

## Abstract

To deconstruct the multiple levels of risk factors for Clostridioides difficile infection, using multilevel models (MLMs) accounting for patient movement.

Case-control study of patients hospitalized in three acute care Delaware hospitals, December 2019–December 2023.

Cases were patients aged ≥18 years who tested positive for hospital-onset C. difficile infection. Controls were patients aged ≥18 years hospitalized more than 72 hours, who did not test positive for C. difficile infection.

Hierarchical and cross-classified MLMs were used to calculate odds of C. difficile infection based on patient-level risk factors and to evaluate the variation in odds of infection attributable to environmental risk factors using the hospital unit(s) a patient was assigned to during hospitalization.

Our study included 1,223 patients (249 cases, 974 controls). In both models, greater odds of infection were associated with antibiotic exposure [adjusted odds ratio (aOR) = 11.20, 95% confidence interval (CI) = 7.19, 17.40; aOR = 12.80, 95% CI = 8.46, 19.40 for hierarchical and cross-classified models respectively] and health insurance (aOR = 1.74, 95% CI = 1.12, 2.68; aOR = 1.62, 95% CI = 1.03, 2.53; public vs. private). Median odds ratios (MOR) for both models indicated greater relevance of between-unit heterogeneity in the outcome than health insurance but less than antibiotic exposure (MOR = 1.83, 95% CI = 1.56, 2.30 and 2.71 95% CI = 2.10, 4.06).

Using multilevel methods accounting for patient movement, we found that while antibiotic use is the most important risk factor in patients that developed C. difficile infection, environmental risk factors are additionally important and should be considered in research involving hospitalized patients and healthcare-associated infections.

## Full-text entities

- **Diseases:** C. difficile infection (MESH:D003015), infection (MESH:D007239)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12926340/full.md

## Figures

6 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12926340/full.md

## References

23 references — full list in the complete paper: https://tomesphere.com/paper/PMC12926340/full.md

---
Source: https://tomesphere.com/paper/PMC12926340