Toxic Exposures Among Young Children One Year into the COVID-19 Pandemic: A Retrospective Review of Three San Francisco Bay Area Emergency Departments

Gabriel Devlin, Newton Addo, Ashkon Shaahinfar, Carol C. Chen, Jacqueline Grupp‐Phelan, Aaron E. KornblithJournal of Emergency Medicine 2022 · DOI

The Journal of Emergency Medicine, the publication venue for this study.

Abstract

This retrospective study investigates the impact of COVID-19 "shelter-in-place" (SIP) orders on pediatric toxic exposures in the San Francisco Bay Area. Analyzing 140,648 emergency department (ED) visits by children aged 0–5 years over a five-year period, the study found that while the absolute number of toxic exposures remained stable, the relative incidence nearly doubled due to a significant decrease in overall ED volume. Notably, the odds of cannabis ingestion nearly tripled during the pandemic year (OR = 2.70), whereas corrosive substance exposures decreased (OR = 0.40). Clinical severity, including ICU admission rates and length of stay, showed no significant change. These findings highlight a shift in household hazards during the pandemic, particularly the rising risk of cannabis exposure in young children.

Introduction

The onset of the COVID-19 pandemic in March 2020 led to unprecedented "shelter-in-place" (SIP) orders globally. For families with young children, this meant the closure of daycares and schools, resulting in children spending nearly all their time at home. Public health experts initially hypothesized that this increased "exposure time" to household chemicals and medications would lead to a surge in accidental ingestions. While national data suggested some increases, regional granularity was often lacking. This study examines the San Francisco Bay Area—a region with some of the strictest and longest-running SIP policies in the United States—to determine if these local restrictions influenced the rates or severity of pediatric toxic exposures.

Methods

Researchers conducted a retrospective cross-sectional study using electronic medical record (EMR) data from a tertiary care hospital system in San Francisco, encompassing two pediatric EDs and one general ED. The study population included children aged 0–5 years.

The analysis covered a five-year window from March 16, 2016, to March 15, 2021, with the "pandemic era" defined as the final year (March 16, 2020, to March 15, 2021). Patients were identified using ICD-10 diagnosis codes for poisoning, toxic effects, or drug intoxication, categorized using the CDC Injury Mortality Diagnosis Matrix. Logistic regression was employed to calculate odds ratios (OR) for specific exposure types, and the Kruskal-Wallis test was used to compare lengths of stay, comparing the pandemic year against the aggregate of the four preceding years.

Results

The Denominator Effect and Relative Incidence

The study identified a total of 770 confirmed cases of toxic exposure out of 140,648 ED encounters. A primary finding was the "denominator effect": while the rate of toxic exposures per 1,000 ED visits nearly doubled from 4.99 to 9.79 during the pandemic, the absolute number of poisonings remained flat. This statistical rise was driven by a 55% plummet in overall ED volume as parents avoided hospitals for non-emergent issues.

Shifting Exposure Profiles

The composition of toxic exposures shifted significantly during the pandemic:

  • Cannabis Ingestion: The odds of a child presenting with cannabis ingestion were 2.7 times higher during the pandemic than in previous years (OR = 2.70; 95% CI 1.60–4.49). This likely reflects the combined impact of SIP orders and the increasing availability of cannabis products, such as edibles, in the home.
  • Corrosive Substances: Conversely, the odds of corrosive substance exposures (e.g., bleach, disinfectants) decreased (OR = 0.40; 95% CI 0.15–0.88). While public reports suggested increased household use of cleaners, fewer incidents were severe enough to require emergency intervention.

Clinical Severity

Despite the stressors of the pandemic, the clinical severity of cases remained consistent. No significant differences were detected in triage acuity (7% vs. 8% high acuity), hospital admission rates (31% vs. 30%), or ICU admissions (11% vs. 9%). Length of stay also remained stable across the study periods.

Discussion

The findings suggest that while the pandemic did not cause a surge in the total number of pediatric poisonings requiring emergency care, it significantly altered the types of substances children were exposed to. The tripling of cannabis ingestion cases highlights a critical public health concern as legalization and home storage of edibles become more common.

Limitations and Future Directions

The study is limited by its reliance on ED visit data, which likely represents only the "tip of the iceberg." Mild or asymptomatic exposures managed at home or via poison control calls without hospital transfer were not captured. Additionally, the findings from a single San Francisco hospital system may not be generalizable to regions with different lockdown stringencies. Future research should integrate data from the California Poison Control System (CPCS) to provide a more comprehensive view of household safety during public health crises.

Conclusion

One year into the COVID-19 pandemic, the absolute number of pediatric toxic exposures in the San Francisco Bay Area remained stable, though the relative risk of cannabis ingestion increased significantly. These results underscore the need for targeted public health messaging regarding the safe storage of cannabis products in households with young children, especially during periods of increased time spent at home.

Summary updated

BibTeX
@article{devlin2022toxic,
  author = {Gabriel Devlin and Newton Addo and Ashkon Shaahinfar and Carol C. Chen and Jacqueline Grupp‐Phelan and Aaron E. Kornblith},
  title = {{Toxic Exposures Among Young Children One Year into the COVID-19 Pandemic: A Retrospective Review of Three San Francisco Bay Area Emergency Departments}},
  journal = {Journal of Emergency Medicine},
  year = {2022},
  doi = {10.1016/j.jemermed.2022.09.035},
}
Figure