Smell tests could be the answer to 'moonshot'

This week, new research revealed that loss of sense of smell could be a clearer sign of coronavirus than a cough. A study by University College London (UCL) of 590 people who lost their sense of smell or taste earlier in the year found 80pc had coronavirus antibodies.

The results of this study aren’t surprising. Back in April, we published research in the journal Nature Medicine from over 2 million app users that showed a loss of sense of smell is far more predictive than cough or shortness of breath in establishing definite cases of coronavirus. Although antibodies aren’t as reliable as swab testing, this study is important in reaffirming what we already knew; anosmia is the cardinal symptom of coronavirus.

The data to support this keeps on growing. Our latest figures show that across all age groups, 55 per cent of adults aged 18-65 who have a positive swab test for coronavirus have a loss of sense of smell or taste (called anosmia). In fact, if you have anosmia, your odds of testing positive for Covid are around ten times higher than having a fever or cough. 

A loss of smell is a good predictive factor for coronavirus because it doesn’t occur commonly in any other illness. Meanwhile, a cough, shortness of breath and fever are non-specific symptoms that happen with many other common viruses and infections. We did a study with 400 people from our twin cohort that showed 90 per cent of people with a loss of smell also had antibodies for Covid. No one knows for sure why anosmia happens, although it could be caused by nerve damage to the receptors in the lower part of the brain, where it meets the nose. The virus might be trapped there, and with an immune reaction, the whole area gets irritated. 

Given the prevalence of the symptom, could smell testing be the answer to Boris Johnson's moonshot plan? It would certainly be much more effective than temperature testing. While a fever only lasts on average for two days, anosmia lasts for ten days so there’s a higher chance of it being picked up. For some people, it can also be the only symptom; in our data, 16 per cent of people had a loss of smell and no other classic signs of coronavirus. 

Everyone has such a wide variety in normal temperature; while the cut-off point for a high temperature is 38 degrees, some people might be feverish but find their temperature never gets that high. Plus, if you have a fever you're much less likely to be travelling. I think the temperature tests are lulling people into a false sense of security. 

The Department of Education recently said that too many children and school staff are getting tested for Covid-19 despite having no symptoms. A much better option would be to test people using a triage format. You could smell test children coming into schools. If they report symptoms, then isolate them and get them tested. Children and the elderly are much less likely to report anosmia, so we should also be getting people to test elderly relatives or children themselves rather than waiting for them to say: 'I can’t smell very well.'

The same could, in theory, be done in airports, hospitals and shopping centres. At the entrance to a hospital clinic, for example, you could be asked to wash your hands, have your temperature taken, and you could have items to smell. A major caveat is that you don’t want to prejudice against people who have long-Covid or older people who may have similar longterm problems. However, if you stuck to smell testing children, who are much less likely to report the symptom, and have short term problems, I don’t see a reason for it not to work. 

Some schools and airports have been trialling smell tests in Asia. Meanwhile, researchers in Boston have developed an at-home smell test that was trialled on 400 patients across three hospitals in the local area. I also know about ‘scratch and sniff’ tests which can be sent to your home. But I think mostly it can be done at home by seeing if you can smell common household items, such as cinnamon, rosemary leaves or coffee.

So while the study isn’t entirely new, I hope it helps to educate people further about the importance of anosmia. In my opinion, it's the defining symptom of coronavirus. 

Professor Tim Spector of King's College London leads the Covid Symptom Study, the world’s largest citizen science project of over 4 million users which allows people to report their symptoms and track the spread of the virus in their area. You can download it on the app store, or visit the website here