Enlarge /. A drive-through coronavirus test station.
A dry cough. Loss of smell. Diarrhea. Fever. All of these symptoms were considered to be possible symptoms associated with SARS-CoV-2 infection, along with the complete absence of symptoms. In the absence of adequate testing capacity, many areas in the United States are forced to assign their limited tests only to those who appear to have COVID-19 symptoms. However, given the difficulty in determining which symptoms actually indicate a likely infection, these decisions are difficult to make.
The confusing array of symptoms also raises questions about why people react so differently to the same virus.
Finding out what's going on in the middle of a pandemic is an incredible challenge. We'll look at some preliminary reports of a way to do this – not because the results are likely to continue when more research comes in, but because they show some of the ways researchers use to try to understand the virus's infection .
Symptom or coincidence?
Currently, the CDC website lists a variety of symptoms associated with COVID-19. Some of these are what you would expect for a viral infection of the lungs: fever and chills, cough, shortness of breath and sore throat. But there are also some less obvious ones, such as headaches, muscle aches and loss of smell.
Lists like this are usually created by aggregating medical reports, as doctors record and update a person's symptoms as soon as they are included and treated. However, the lack of testing poses significant problems for this effort. First, we have difficulty understanding how many people have been infected without needing medical care. The pandemic also coincided with the flu season and seasonal allergies, which can lead to overlapping symptoms.
Finally, the list of symptoms is generally a product of the patient's own memory, as asked to describe the beginning of the problem. Memories can be problematic because the need for medical care can improve the retrieval of symptoms that might otherwise be ignored. The broad awareness of lists of symptoms like that of the CDC can serve the same purpose.
To make matters worse, some problems affect a relatively small group of infected people. Various reports have involved SARS-CoV-2 in the formation of blood clots, possibly due to its infection of the lining of blood vessels. The same applies to the gastrointestinal symptoms. And the receptor that the virus uses to penetrate cells is also in the kidneys, which could possibly explain why a number of hospitalized patients needed dialysis. Why some patients are prone to severe symptoms while others have an asymptomatic infection remains unclear.
A lot of laboratory work is ultimately required to find out what's going on with some of them. We need to find out whether the virus can multiply in kidney cells and, for example, examine the type of kidney damage. Another group of researchers has been looking for ways to understand the occurrence of COVID-19 symptoms in real time.
What happens if?
Two recent drafts described the first results of a collaboration between health researchers at Harvard and King’s College London. They worked with an application development company to put together a simple mobile device application called COVID Symptom Tracker that asks its users a series of questions every day. These focus on known symptoms of COVID-19 (and can be updated as the list expands) and all test results and treatments received.
This approach has some drawbacks. The users are self-selected and have smartphones, which probably means a younger population. And the app only asks for the symptoms that the developers have activated. However, the approach swaps these limitations for some significant benefits. The first is simply scaling. The app had 2.2 million users between the United States and Great Britain before the end of March. The second is that it does not suffer from recall bias – people enter their symptoms as soon as they appear before they know they are associated with a positive diagnosis. Later, early symptoms and even treatments may be correlated with patient outcomes.
How does it work? A draft describing the application contains an initial analysis of the users. It turned out that people were usually tested after reporting cough and / or fatigue, but these weren't really linked to the positive result of the test. The same was true if someone had diarrhea, but no other symptoms. Instead, positive diagnoses were made if cough and / or fatigue were associated with diarrhea or loss of smell. In fact, the results indicated that loss of smell was more common than fever among those who eventually tested positive for the virus.
Last week, the team published a follow-up draft examining the characteristics associated with these symptoms. This is done using a group of twins in the UK who volunteered to participate in health research. Some of them (just over 2,500) used the application, and the researchers tried to find out if genetics could affect people's symptoms.
The problem here is that not enough of them have been diagnosed to do any kind of analysis. So the researchers did not correlate with actual SARS-CoV-2 infections. instead, they correlated it with a prediction of a positive diagnosis, as stated in their earlier draft manuscript. That is clearly the biggest weakness of this work. The second is that it covered an extremely short period: March 25th to April 3rd.
Given these precautions, you should take the results with an entire salt desert. However, there are some interesting potential results that seem worth pursuing with longer-term tracking. They suggest that there is a significant genetic component that is likely to be genetically obtained for a COVID-19 diagnosis with a probable 50 percent value (although the error range was between 30 and 80 percent). In addition, many symptoms appeared to be genetic, including fever, fatigue, loss of smell, and diarrhea. Others, such as cough, chest pain, and abdominal pain, were not.
Although this study will not change our understanding of the current form, it does offer an interesting model of how a higher quality study would work – this is a possible reason to even publicly publish a draft. And it is clear that at some point this study could be what we are looking for as there is no reason why we could not test every single participant and get a clearer picture of what is going on. In the longer term, instead of relying on twins, we would also like to conduct genome-wide studies with unrelated people, which could significantly expand the study population.
Although this study shouldn't be much more than an indication of what is possible, something clearly affects the body's response to this virus in different people. Genetics is a reasonable candidate for one of the influences there.