From NIH Director Francis Collins' Blog:
In the current study researchers from King's College London, Mass General Boston, and others studied 4182 COVID patients over the course of their illness as they logged their symptoms using a Smartphone app.
Using this information they developed a model to predict which patients would go on to develop "Long COVID."
Those most likely to develop "Long COVID" were older, female, and those who experienced five or more symptoms within the first week of illness . The latter variable proved to be the strongest predictor; interestingly, the nature and order of the symptoms didn't matter.
The model accurately predicted 70 percent of cases and was validated on an additional group of 2472 COVID patients.
"Real-time tracking of self-reported symptoms to predict potential COVID-19," Menni C, et al
Nature Medicine 2020 July 26.
"Attributes and predictors of long COVID," Sudre CH, et al, Nature Medicine 2021 March 10.