A British study used a machine learning and symptom tracking application to group the most common symptoms of COVID-19. The researchers also assessed the likelihood that each type of symptom would lead to a serious case.
A tracking app shows its value
It is estimated that more than 4 million people use the Covid Symptom Tracker application. The app is intended to serve as a tracker to see how coronavirus cases are changing across different communities, and also to learn about the symptoms associated with COVID-19.
Any information can save lives, and we have already seen that the virus can be extremely unpredictable. Symptoms can range from diarrhea and confusion to chest pain and extreme fatigue, but we’re still not sure how common these symptoms are and how likely they are to indicate that something bad is coming. Knowing whether someone is likely to need ventilation a couple of days in advance could make a difference, both for hospitals (which can prepare accordingly) and for patients themselves.
In the study, the team used a machine learning algorithm to explore whether symptoms can be clustered. Overall, the team obtained data from 1,653 users who tested positive and regularly noted their symptoms and health situation. Of these, 383 went to the hospital at least once, and 107 required additional oxygen or ventilation. According to the researchers, the symptoms can be grouped together and can be used to assess the likelihood of a severe case.
The key is to take several symptoms together.
“Since no symptoms can predict the severity of the disease or the need for dedicated medical care in COVID-19, we asked whether documenting time series of symptoms in the first few days would inform the outcome,” the authors write.
The six categories of coronavirus symptoms
After the researchers collected the patient data, they entered it into an unsupervised machine learning algorithm that grouped the symptom groups. The six groups of symptoms the researchers found are:
- Group 1: symptoms in much of the upper respiratory tract (especially persistent cough). Muscle pain was also present. About 1.5% of patients in this group required respiratory assistance, and 16% made at least one visit to the hospital. This was the most common group.
- Group 2: symptoms in much of the upper respiratory tract, but a higher frequency of skipped meals and higher fever. In this group, 4.4% required respiratory assistance and 17.5% visited the hospital.
- Group 3: gastrointestinal symptoms (such as diarrhea), but surprisingly fewer symptoms. Here, 8.6% required respiratory assistance and 23.6% made at least one trip to the hospital.
- Group 4: early signs of fatigue, severe chest pain, and persistent cough. Out of this group, 8.6% required respiratory assistance and 23.6% went to the hospital.
- Group 5: confusion, severe fatigue and many skipped meals. Out of this group, 9.9% required respiratory assistance and 24.6% visited the hospital.
- Group 6: respiratory distress, which includes shortness of breath and chest pain, in addition to fatigue, confusion, and gastrointestinal problems. Almost 20% required respiratory assistance and 45.5% went to the hospital at least once.
The researchers also note that a higher body mass index correlated with more severe symptoms, as did older age and chronic lung disease. Men were also more likely to report severe symptoms.
The first two groups were associated with milder forms of the disease. The third was quite unusual, as symptoms tended to manifest at the gastrointestinal, rather than the respiratory, level. Then the other groups get progressively worse. Groups 5 and 6 had the highest risk of hospitalization and were more likely to require artificial ventilation. Groups 3 and 4 were also at relatively high risk.
If used widely, this could provide important information to healthcare providers. It could allow patients to be monitored remotely and help predict how many hospital beds and ventilators would be needed in the near future.
The team says the findings could give health care providers a multi-day advance warning of demand for hospital care and respiratory care.
It could also help point patients at risk of serious illness by directing support at home, such as an oxygen meter or visits to nurses. Currently, the authors write, the average time to reach the hospital is 13 days.
However, the study should also be taken with the grain of salt. The sample size only included a few hundred patients for each group, which may not be enough to get an accurate picture of what is happening. The grouping itself was performed using an algorithm simply by association; this is more an indication than a clear classification of symptoms. Finally, the study has not been peer reviewed and is awaiting expert analysis.
However, as we collect more and more data on pandemics, it is increasingly clear that the picture is far from clear, and the data is often confusing or inconclusive. It is this type of classification that could help us better understand the disease and how to prepare local hospitals.
If this study is confirmed, it could save lives.
The study has been published on medRxiv.
