Critical Review: Preliminary indicators of mortality in subjects infected with COVID-19 based on data from China and South Korea

Authors: Shahriar Zehtabchi, MD; Joe Habboushe, MD, MBA 
Reviewed by: Cassidy Dahn, MD; Kyan Askari, MD

This article was published in collaboration with MDCalc. Please see the MDCalc COVID-19 Resource Center for more information, including a critical review of recommended calcs for resource-limited situations and more.

An odds ratio (OR) is a comparison of the odds of an event or outcome in those with a particular risk factor, versus the odds of that outcome or event in those without that particular risk factor. OR is a useful way to estimate how the presence of a risk factor increases the likelihood of an event. For example, age ≥60 years can significantly increase the risk of mortality in patients with community-acquired pneumonia (CAP).1

Figure 1.

Many practicing clinicians have developed a gestalt of prognostic indicators based on experience or are familiar with scores to help prognosticate outcomes for patients with CAP. As a point of reference, the ORs for CAP and COVID-19 (Table 1) are shown together in order to help clinicians put into context different risk factors with magnitudes higher ORs and similar ORs when making clinical decisions for patients with the novel coronavirus.

Table 1: Risk factors associated with mortality in subjects infected with CAP and COVID-19

Image from Gyazo

Figure 2.

To be clear, these figures do NOT compare the absolute mortality rate between CAP and COVID-19 patients with pneumonia. It is also important to note that the data presented here are mainly from the first cohort of COVID-19 patients from China.2, 3 Therefore, this information may not be generalizable to other populations, and is also limited in that it only includes those patients who were tested. Lastly, the characteristics of patients and the behavior of the disease is constantly evolving. Therefore, we emphasize on the “preliminary” nature of the information provided here.

Of note, the prevalence of Diabetes (DM) and cardiovascular disease (CVD) in the cohort of patients reported by the Chinese CDC are notably lower than in U.S. populations, ~2.5% vs. 10.5% (all ages, 13% US adults) and ~2.0% vs. 12.1% (adults with diagnosed heart disease) respectively.2, 3, 4, 5 This could be in part due to missing information. If so it could also indicate that patients with DM and/or CVD in this dataset are generally sicker than the average patient with DM and/or CVD in the U.S., which would inflate the OR.

Figure 3.

In addition, we have listed the prognostic value of some of the lab abnormalities reported in various datasets (Table 2). We encourage the readers to interpret these results with caution as the numbers are derived from datasets in a retrospective manner with a significant percentage of missing data. Therefore, at best, they are suggestive of poor outcomes and their prognostic value has to be validated properly in future studies that are prospective with improved methodology. In addition, most of the ORs have very wide confidence intervals (most likely due to small sample sizes), which further limits the validity of the data. While the original Chinese cohort includes over 44,000 patients, some of the lab values are based on studies of less than 200 patients.

Table 2. Laboratory Values associated with mortality

Image from Gyazo

Examining the existing data highlights the importance of age as the most important prognostic factor. Therefore, we have listed the fatality rates from COVID-19 based on age groups in Table 3. The data indicates more than a triple-fold increase in mortality from COVID-19 infection in patients in age group 60-69 years compared to those aged 50-59 years. The mortality continues to rise significantly in persons aged 70 years and above.

Table 3. Fatality rate in different age groups derived from Chinese CDC2 and South Korea CDC6 datasets

Image from Gyazo