SA3 - Does the Closure of a Safety-Net Hospital Increase the Complexity of Odontogenic Infections at a Receiving Tertiary Center?

Abstract: Hahnemann University Hospital, a 496-bed safety-net hospital in Center City Philadelphia, closed abruptly in September 2019, eliminating emergency surgical access for predominantly uninsured and Medicaid populations. While safety-net closures are increasing nationwide, their downstream impact on surgical disease complexity at receiving institutions has not been quantified. When patients lose proximity to definitive surgical care, delays in treatment may allow odontogenic infections to progress from single-space to multi-space involvement before operative intervention. This investigation used odontogenic infections as a novel tracer condition to determine whether Hahnemann's closure increased infection complexity at the neighboring tertiary oral and maxillofacial surgery center at the Hospital of the University of Pennsylvania.

A retrospective cohort study compared adults admitted to the Penn Oral and Maxillofacial Surgery service with odontogenic infections requiring operative intervention before (September 2016 through September 2019, 37 months) and after closure (September 2019 through September 2022, 36 months). Non-odontogenic infections including medication-related osteonecrosis and radiation-associated cases were excluded, along with trauma, non-operative management, and patients under 18 years. After collapsing same-date billing duplicates, 211 unique surgical encounters were identified (107 pre-closure, 104 post-closure). Fascial space involvement was abstracted from operative notes by dual independent reviewers with consensus resolution of discrepancies, yielding 163 encounters with complete fascial space documentation (78 pre, 85 post; 77% abstraction rate). The primary outcome was mean number of fascial spaces involved per case. Secondary outcomes included multi-space infection prevalence (2 or more anatomic spaces), deep neck space involvement (lateral pharyngeal, retropharyngeal, pretracheal, danger space, or mediastinum), patient demographics, and annualized case volume. Because the post-closure period substantially overlapped with the COVID-19 pandemic, multivariable linear and logistic regression models were constructed adjusting for pandemic period (March 2020 through June 2021) and patient age to isolate the independent effect of hospital closure from pandemic-related changes in healthcare utilization.

Mean fascial spaces were significantly higher post-closure (1.65 ± 1.09 vs 2.16 ± 1.34, P = .009), with multi-space infections increasing from 50.0% to 63.5% (P = .113 unadjusted) and deep neck involvement from 9.0% to 11.8% (P = .745 unadjusted). Critically, annualized volume remained stable at 34.7 cases per year in both periods, indicating that the closure did not simply redistribute patients but rather shifted the severity profile toward more advanced disease at presentation. Multivariable analysis revealed COVID-19 was independently associated with decreased infection complexity (beta = -0.53, P = .048), confirming the pandemic suppressed disease severity and masked the full closure effect. After adjustment, Hahnemann closure was independently associated with 0.85 additional fascial spaces per case (95% CI: 0.37-1.33, P < .001), 3.30-fold increased odds of multi-space infection (95% CI: 1.35-8.06, P = .009), and 3.17-fold increased odds of deep neck space involvement (95% CI: 1.03-9.75, P = .045). Age, sex, and race did not differ between groups.

This study provides the first evidence that safety-net hospital closure independently increases odontogenic infection complexity at a receiving tertiary center without increasing case volume. This pattern is consistent with a delayed access model: loss of a proximal emergency surgical facility increases the interval between infection onset and definitive operative intervention, allowing infections to progress from single-space to multi-space involvement before reaching surgical care. These findings carry direct implications for surgical planning and ICU resource allocation at institutions absorbing displaced populations and establish odontogenic infections as an objective tracer condition for quantifying the downstream impact of healthcare access disruptions on vulnerable communities.


References:


1.Hsia RY, Kellermann AL, Shen YC. Factors Associated With Closures of Emergency Departments in the United States. JAMA. 2011;305(19):1978-1985. doi:10.1001/jama.2011.620.

2.Gams K, Shewale J, Demian N, Khalil K, Banki F. Characteristics, length of stay, and hospital bills associated with severe odontogenic infections in Houston, TX. J Am Dent Assoc. 2017;148(4):221-229. doi:10.1016/j.adaj.2016.11.033.