A History of Psychiatric Hospitalization and Surgically Managed Facial Abscess: An Observed-to-Expected Association Analysis.

Background/Objectives: Severe mental illness is associated with poor oral health, and patients with psychiatric conditions are hospitalized for dental disease more often than the general population. Whether this extends to facial abscesses severe enough to require surgical drainage has not been examined, particularly in United States safety-net populations. We sought to determine whether patients with a history of psychiatric hospitalization are overrepresented among surgically managed facial abscess cases relative to what published general-population rates would predict. Methods: Patients were identified at a single tertiary safety-net center between 22 February 2014 and 1 April 2024 using ICD-10 codes for facial abscess together with Epic SlicerDicer psychiatric diagnosis concepts/groups, and cases were confirmed by individual chart review. Observed cases were defined as patients with documented psychiatric hospitalization and documented surgical management of a facial abscess. A denominator of unique psychiatric inpatients over the same interval was obtained through a separate query. Because that query returned an aggregate count without denominator-level follow-up dates, expected cases were estimated by applying published general-population inpatient facial-abscess rates (base case 2.5 per 100,000 person-years; range 2.0-4.0) to approximate person-time, and a crude observed-to-expected ratio with exact Poisson 95% confidence intervals (CIs) was calculated. Temporal ordering of abscess and psychiatric admission was examined where sufficient timing data were available. Results: Eighty patients met the case definition, against 3.37 expected under the base-case rate, giving a crude observed-to-expected ratio of 23.8 (95% CI 18.9-29.6). The ratio ranged from 14.9 to 29.7 across the reference-rate range. Among the 67 cases with sufficient timing data, 30 (45%) preceded the first psychiatric admission, 1 (1%) occurred during a psychiatric hospitalization, and 36 (54%) followed discharge, so the finding is presented as an association rather than an incidence. The abscess was odontogenic in 45 cases (56%), non-odontogenic in 19 (24%), and unspecified in 16 (20%). Substance use disorders (32.5%) and schizophrenia-spectrum disorders (32.5%) were most frequently represented, with a mean of 2.8 psychiatric admissions per patient. Conclusions: In this single-center safety-net dataset, patients with a documented history of psychiatric hospitalization were substantially overrepresented among surgically managed facial abscess cases relative to published population rates. Because the analysis is a crude observed-to-expected comparison without dynamic denominator-level person-time or demographic standardization, and because a substantial minority of abscesses preceded the psychiatric admission, the ratio reflects the combined burden of psychiatric hospitalization and its associated exposures rather than an independent causal effect. Patient-level linked cohort studies are needed to estimate the magnitude of risk reliably.