The coronavirus disease 2019 (COVID-19) has spread worldwide, placing a heavy burden on limited healthcare resources [1]. Although the Omicron variant is significantly less pathogenic, its enhanced infectivity has resulted in an increase in infections and deaths [2]. Given the wide range of clinical manifestations in COVID-19 patients [3], it's crucial to explore simple and effective indicators to help identify high-risk patients, rationally allocate medical resources, and timely guide clinical intervention. These measures are important to improve prognosis and reduce mortality.

Previous studies have shown that autonomic dysfunction is associated with poor prognosis, and heart rate variability (HRV) parameters have been identified as predictors of mortality for many diseases [[4], [5], [6]]. Patients with COVID-19 may experience autonomic dysfunction, manifested by increased sympathetic activity, decreased vagus function, and low HRV [7,8]. At present, a large number of studies have investigated the prediction model of mortality of COVID-19 patients, but most of these focused on the predictive value of clinical symptoms, vital signs, blood indicators, and pulmonary imaging indicators [9,10], with few studies exploring the association between HRV and mortality in COVID-19 patients.

HRV is a commonly used indicator of autonomic nerve function, which can be obtained by converting R-R intervals from electrocardiogram (ECG) recordings [11]. This is a simple and low-cost method with reliable data. Previous studies have demonstrated a strong association between ultra-short HRV (US-HRV) measured by 10-s ECG recordings and HRV by longer ECG recordings [4,12,13]. The purpose of this study was to investigate the association between US-HRV and the prognosis of patients with COVID-19, and to construct two prediction models applicable to guide clinical practice in different scenarios.