Abstract
BACKGROUND: Older adults (> 65 years) residing in long-term care facilities (LTCFs) are at elevated risk of severe outcomes from respiratory infections. Infections often remain undetected or present atypically in this population, leading to underdiagnosis. Our study aimed to estimate the respiratory virus infection burden, independent of symptom presentation, among older adults in Swedish LTCFs in the post-pandemic period (2021-2024). METHODS: We leveraged capillary blood samples and coupled national registry data from 1622 LTCF residents (median age = 87). A multiplex platform was used to quantify antigen-specific IgG and IgM responses to RSV (pre-/post-F, strain A-specific G-protein), influenza-A (H1N1 and H3N2 HA), influenza-B (HA) and SARS-CoV-2 (spike). Linear mixed-effects models were used to demonstrate the dynamics of antibody levels over time, adjusted for age, sex and comorbidities. RESULTS: RSV-specific antibody responses peaked in spring 2022 (p < 0.001), suggesting an impact of relaxed COVID-19-related restrictions on RSV exposure at LTCFs. RSV-specific antibodies subsequently declined over time until an increase during autumn 2023 (p < 0.001). Geographic variation in pre-F antibody levels suggested localised RSV outbreaks. The total estimated RSV burden at LTCFs was markedly higher than official reports of the Swedish Public Health Agency. Influenza antibody dynamics reflected seasonal trends and were strongly influenced by annual vaccination. A random forest classifier incorporating serological profiles with demographics, location and comorbidities significantly outperformed a model without serological data (AUC-ROC = 0.67 vs. 0.58), although discriminatory performance remained modest. Higher levels of RSV pre-F antibodies in autumn 2021 were associated with increased one-year mortality in logistic regression (OR = 1.43, p = 0.024). Exploratory survival analysis indicated a trend that elevated levels of RSV pre-F antibodies during low population immunity may confer a transiently elevated early hazard of death, although this did not reach statistical significance (HR = 4.50, p = 0.087). CONCLUSIONS: We observed substantial respiratory virus circulation among older adults in Swedish LTCFs and show that RSV burden is under-reported. The results highlight a need for further research into the role of RSV pre-F antibody levels in preventing severe outcomes, potentially via vaccination of LTCF residents. Our scalable serological surveillance system is a valuable approach to detect respiratory infections in LTCFs, independent of symptom presentation or healthcare-seeking behaviour.