Structural Reconfiguration of the Time-Valid Cohort and Stage-Specific Reversal of Prehospital Time-Outcome...

Background and Objectives: During the COVID-19 pandemic, worsening outcomes after out-of-hospital cardiac arrest (OHCA) have been widely reported and are often attributed to prolonged prehospital system delays. However, little attention has been paid to whether the population of patients with analyzable prehospital time data-the time-valid cohort-itself changed during the pandemic, or to how such changes may have influenced the observed association between prehospital time and outcomes. To examine structural changes in the time-valid OHCA cohort across pandemic phases, and to evaluate phase-specific associations between call-to-emergency department (ER) time and neurological outcomes. Materials and Methods: We conducted a nationwide retrospective observational study using a Korean OHCA registry from 2016 to 2022 (n = 217,356). Patients with logically consistent prehospital time intervals from arrest recognition to ER arrival were defined as the time-valid cohort (n = 62,240). Pandemic phases were categorized as pre-pandemic (2016-2019), early pandemic (2020), and prolonged pandemic (2021-2022). Changes in cohort composition were assessed descriptively and visually. Associations between call-to-ER time (per 10-min increase) and good neurological outcome (Cerebral Performance Category 1-2) were evaluated using phase-stratified logistic regression models adjusted for age, sex, and initial rhythm. Results: The time-valid cohort consisted exclusively of witnessed arrests throughout the study period. As the pandemic progressed, the cohort became older (median age increased from 70 to 72 years), and the proportion of shockable rhythm declined from 21.7% in the pre-pandemic period to 17.5% in the prolonged pandemic period. The proportion of good neurological outcomes decreased from 9.3% to 6.8%. Before the pandemic, longer call-to-ER time was associated with a lower likelihood of a good neurological outcome (odds ratio [OR] per 10-min increase, 0.85; 95% confidence interval [CI], 0.83-0.88). This association was attenuated during the early pandemic phase (OR, 0.95; 95% CI, 0.91-1.00) and reversed during the prolonged pandemic phase (OR, 1.07; 95% CI, 1.04-1.10). Conclusions: Changes in the association between prehospital time and neurological outcome during the COVID-19 pandemic cannot be interpreted as the effect of system delay alone. Instead, these findings should be understood in the context of substantial structural reconfiguration of the time-valid OHCA cohort, which became progressively older and physiologically less favorable across the pandemic's phases. Consideration of cohort structure is essential when interpreting prehospital time-outcome relationships during large-scale system disruptions.