The somatic symptom burden of patients with medically unexplained oral symptoms across the COVID-19 pandemic: a retrospective cross-sectional study
Background To determine whether COVID-19 pandemic periods were associated with the general somatic symptom burden of patients with medically unexplained oral symptoms (MUOS) and to identify patient-level and diagnostic correlates. Methods This retrospective cross-sectional study included 3,542 adults presenting for initial consultation at a psychosomatic dentistry clinic from April 2017 through September 2025. Somatic burden was assessed using the Somatic Symptom Scale-8 (SSS-8), and depressive symptoms using the Zung Self-Rating Depression Scale (SDS). World Health Organization declarations defined three periods: before the pandemic, during the pandemic, and after the end of the Public Health Emergency of International Concern (PHEIC). Group differences and adjusted associations with SSS-8 scores were examined using analysis of variance, multiplicity-adjusted pairwise tests, and multivariable linear regression with categorical period indicators, without and with SDS adjustment. Results The mean SSS-8 scores were 9.04 (95% CI, 8.73–9.35) in the pre-pandemic period, 9.45 (9.11–9.79) during the pandemic, and 9.63 (9.23–10.04) in the post-PHEIC period. The overall ANOVA was statistically significant ( p = 0.047), but no pairwise comparison remained significant after false-discovery-rate correction. Without SDS adjustment, post-PHEIC scores were higher than pre-pandemic scores (b = 0.66; 95% CI, 0.17–1.15; p = 0.009; joint period p = 0.028). Period estimates were smaller after adding SDS (joint period p = 0.450). In this model, SDS was the strongest correlate (standardized β = 0.542, p < 0.001); psychiatric history, burning mouth syndrome, persistent idiopathic facial pain, and longer illness duration were associated with higher burden, whereas oral cenesthopathy was associated with lower burden. The adjusted R² was 0.329. Conclusions In this sample of patients with MUOS, observed differences in somatic symptom burden across calendar-period samples were small and depended on SDS adjustment. These comparisons cannot establish causal pandemic effects, absence of such effects, or mediation by depressive symptoms. Longitudinal studies incorporating individual pandemic-related exposures are needed.
Authors
- Trang Thi Huyen Tu (ORCID: https://orcid.org/0000-0002-6062-4603)
- Akira Toyofuku (ORCID: https://orcid.org/0000-0002-9498-3014)
- Ko Matsudaira
- Takayuki Suga (ORCID: https://orcid.org/0000-0001-7907-2127)
- Zhenyan Liu
- Motoko Watanabe
Publication Details
- Journal
- BioPsychoSocial Medicine
- Published
- 2026-11-19
- DOI
- https://doi.org/10.1186/s13030-026-00378-8
- Primary Topic
- Psychosomatic Disorders and Their Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00