JMIR Publications

ABSTRACT

Background:

The coronavirus disease 2019 (COVID-19) pandemic continues to ravage and burden hospitals around the world. The epidemic started in Wuhan, China was recognized by the World Health Organisation (WHO) as an international public health emergency and declared the outbreak a pandemic in March 2020. Since then, the scale of disruptions caused by the COVID-19 pandemic is having an unparalleled effect in recent history.

Objective:

With increasing total hospitalization and Intensive Care Unit (ICU) admissions, a better understanding of features related to patients with COVID-19 could help healthcare workers stratify patients based on the risk of developing more severe diseases. Using predictive models, we strive to select features that are the most relevant to more severe COVID-19 diseases.

Methods:

Over 3 million participants have reported their potential symptoms of COVID-19 along with their comorbidities, demographics and symptoms on a smartphone-based app. Of the >10,000 who were tested positive, we leveraged Elastic Net regularized binary classifier to derive predictors that are most correlated with users having severe enough disease to seek treatment in a hospital setting. We then analyzed such features in relation to age and other demographics and their longitudinal trend.

Results:

The most predictive features found include fever, the use of immunosuppressant medication, mobility aid, shortness of breath, and fatigue. Such features are age-related and some are manifesting disproportionally in minority populations.

Conclusions:

Predictors selected from the predictive models can be used to stratify patients into groups that may need additional attention than others. It could help health care workers to devote valuable resources to prevent the escalation of diseases in vulnerable populations.