The data taken between January 2020 and April 2021 consisted of 486,149 people with prior infection, and 1.9 million people with no indication of coronavirus infection after matching for other clinical diagnoses.
Using data from patients that had not been admitted to hospital, the team of researchers were able to identify the three categories of distinct symptoms.
Anuradhaa Subramanian, research fellow at the Institute of Applied Health Research, University of Birmingham and lead author of the paper, said: “Our data analyses of risk factors are of particular interest because it helps us to consider what could potentially be causing or contributing to long Covid.”
She added: “Women are, for example, more likely to experience autoimmune diseases. Seeing the increased likelihood of women having long Covid in our study increases our interest in investigating whether autoimmunity or other causes may explain the increased risk in women.