A study analyzing data from 4,182 individuals with COVID-19 who used the COVID Symptom Study app found that 13.3% reported symptoms lasting ≥28 days, 4.5% for ≥8 weeks, and 2.3% for ≥12 weeks. Long COVID was characterized by fatigue, headache, dyspnea, and anosmia, and was more common in older adults, women, and those with higher BMI. Experiencing more than five symptoms in the first week of illness was strongly associated with long COVID (OR = 3.53). A predictive model using data from the first week of illness achieved an AUC-ROC of 76%, with similar results in an independent sample. The model could identify individuals at risk of long COVID for prevention or treatment trials and for planning rehabilitation services. Long COVID was associated with multisystem symptoms, and individuals with long COVID were more likely to require hospital assessment. Asthma was the only preexisting condition significantly associated with long COVID. The study also found that early symptom patterns, such as fatigue, headache, and dyspnea, were predictive of long COVID. A simplified model using age, sex, and number of symptoms in the first week achieved an AUC-ROC of 76.7%. The study highlights the importance of early identification of long COVID to guide healthcare interventions and resource allocation. The findings were validated in an independent dataset, and the study emphasizes the need for further research into the long-term effects of COVID-19, including cardiac and neurological complications. The study also notes the limitations of the data, including the reliance on app users and potential biases in symptom reporting. The results suggest that long COVID is a significant public health issue, with a considerable proportion of symptomatic individuals experiencing prolonged symptoms. The study provides a predictive model that could be used in clinical settings to identify at-risk individuals and support targeted interventions.A study analyzing data from 4,182 individuals with COVID-19 who used the COVID Symptom Study app found that 13.3% reported symptoms lasting ≥28 days, 4.5% for ≥8 weeks, and 2.3% for ≥12 weeks. Long COVID was characterized by fatigue, headache, dyspnea, and anosmia, and was more common in older adults, women, and those with higher BMI. Experiencing more than five symptoms in the first week of illness was strongly associated with long COVID (OR = 3.53). A predictive model using data from the first week of illness achieved an AUC-ROC of 76%, with similar results in an independent sample. The model could identify individuals at risk of long COVID for prevention or treatment trials and for planning rehabilitation services. Long COVID was associated with multisystem symptoms, and individuals with long COVID were more likely to require hospital assessment. Asthma was the only preexisting condition significantly associated with long COVID. The study also found that early symptom patterns, such as fatigue, headache, and dyspnea, were predictive of long COVID. A simplified model using age, sex, and number of symptoms in the first week achieved an AUC-ROC of 76.7%. The study highlights the importance of early identification of long COVID to guide healthcare interventions and resource allocation. The findings were validated in an independent dataset, and the study emphasizes the need for further research into the long-term effects of COVID-19, including cardiac and neurological complications. The study also notes the limitations of the data, including the reliance on app users and potential biases in symptom reporting. The results suggest that long COVID is a significant public health issue, with a considerable proportion of symptomatic individuals experiencing prolonged symptoms. The study provides a predictive model that could be used in clinical settings to identify at-risk individuals and support targeted interventions.