Shifting viral aetiology, codetection and age-specific mortality in severe acute respiratory infection in Brazil, 2022–2025: a nationwide surveillance study

BACKGROUND: Respiratory viral aetiology and codetection among severe acute respiratory infection (SARI) notifications have evolved post-COVID-19. We examined temporal patterns, age heterogeneity, and severe outcome associations in Brazil. METHODS: We analysed nationwide SIVEP-Gripe SARI records (2022-2025; extracted July 2026) with RT-PCR viral detection. Seven mutually exclusive categories were evaluated using robust Poisson regression to estimate adjusted risk ratios (aRRs) for SARI-attributed mortality and a severe composite outcome (ICU admission, invasive ventilation, or death). RESULTS: Among 447,107 virus-positive records, 7.2% involved codetections and 25.4% included RSV. Isolated SARS-CoV-2 declined from 72.4% (2022) to 6.8% (2025), while codetections rose from 3.0% to 11.2%. Compared to isolated RSV, mortality was higher for codetection without RSV (aRR 1.80, 95% CI: 1.67-1.94), isolated SARS-CoV-2 (1.62, 1.54-1.70), and codetection with RSV (1.40, 1.27-1.55). Associations varied by age ($p < .001$). Among RSV-positive cases, codetection conferred increased mortality risk (aRR 1.46, 1.32-1.61). Excluding isolated SARS-CoV-2 attenuated the 2025-versus-2022 codetection prevalence ratio from 1.85 to 1.10. CONCLUSIONS: Viral detections shifted towards greater heterogeneity, strongly driven by changing surveillance composition. Varying testing indications and panels limit causal inference. Integrated, age-sensitive surveillance with standardized diagnostic reporting is essential; surveillance proportions do not reflect population incidence.