The data, which have not been peer reviewed, showed six groups of symptoms, each representing a strain of the coronavirus.
The researchers then identified the group of symptoms that made the patients seriously ill – enough to hospitalize them, the media said.
The clusters vary from mild to potentially fatal.
Symptoms of the mildest strains range from flu-like symptoms without fever to flu-like symptoms with fever and diarrhea.
The three “severe” clusters, reported the Daily Mail, each have a symptom that helps doctors identify tension: fatigue, confusion and, lastly, stomach pain and breathing problems.
With their results, the scientists developed a model that uses age, gender, body mass index and pre-existing health conditions to predict who is at risk for hospitalization, the newspaper reported.
Patients who are more likely to suffer from severe strains are older, overweight and have pre-existing conditions such as diabetes and heart disease.
The prediction tool would be particularly useful in the fall, when public health experts expect another wave of COVID-19 to travel the world again, said lead researcher Dr. Claire Steves.
Doctors could give a warning to patients who may be suffering from the more serious strains and who need to be hospitalized.
“If you can predict these vulnerable people sooner,” said Steves, “you have time to provide support and early intervention to reduce hospitalizations.”