# Rethinking emergency risk assessment: A single-center look at shock index and its variants in COVID-19

## TL;DR

This study examines the usefulness of the Shock Index and its variants in predicting outcomes for moderate COVID-19 patients, finding limited standalone value.

## Contribution

The study evaluates the Shock Index's variants in a moderate COVID-19 cohort, highlighting their limited predictive power and suggesting the need for multivariable models.

## Key findings

- Most shock index variants showed low discriminatory power for mortality and ICU admission.
- Estimated shock index and heart rate–adjusted shock index had higher AUCs for mortality prediction.
- Combining SI with other clinical parameters may improve predictive accuracy but lacks generalizability.

## Abstract

The Shock Index (SI) is a validated prognostic tool in conditions such as severe trauma and obstetric hemorrhage. During the COVID-19 pandemic, it was used to identify patients at higher risk of clinical deterioration, but results have been inconsistent. This study aimed to evaluate the prognostic value of the SI and its variants in predicting mortality, need for mechanical ventilation, and hospital length of stay in patients with moderate COVID-19.

This longitudinal analytical observational study was conducted at a high-complexity hospital in southwestern Colombia and included adults over 18 years of age with moderate COVID-19 treated between 2020 and 2022, using data from the institutional RECOVID registry. A total of 283 patients were analyzed (median age: 61 years; 58.7% male), with cardiovascular and renal comorbidities being predominant. On admission, vital signs were stable (NEWS2: 3.0; shock index: 0.7). ICU admission was required in 29.3% of cases, and overall mortality was 12%. ROC curves and diagnostic accuracy parameters were used to assess the discriminatory ability of the SI and its variants. Most SI variants showed low discriminatory power (AUC < 0.58), except for the estimated shock index and the heart rate–adjusted shock index for mortality, with AUCs of 0.70 and 0.68, respectively, and positive predictive values exceeding 93%.

Early identification of patients at risk for complications in moderate COVID-19 is essential for optimizing hospital resources. The shock index and its variants showed limited utility as standalone predictors for mortality, ICU admission, and hospital length of stay. Combining SI with other clinical parameters may offer some benefit, but heterogeneity limits generalizability. Future studies should develop and prospectively validate multivariable models integrating clinical, laboratory, and imaging biomarkers.