Covid-19 – who should we swab?

25th March 2020, Dr Chee L Khoo

They tell us that patients need to have 1) returned from overseas (any country will do now) OR been in contact with someone confirmed to have coronavirus infection AND 2) Flu symptoms with FEVER before we swab for Covid-19. Otherwise, we just reassure the patient that no they can’t have the swab and yes, they probably don’t have coronavirus infection. We are supposed to be the key frontline health workers in this battle in the pandemic. Have we been given all the information to do our job? Is the information provided correct or is it a sanitised version? Where is the evidence?

There are still a lot of unknowns out there in relation to this evolving pandemic. Let’s look at the many FAQs about this virus:

Selective swabbing – are we missing many cases?

Asymptomatic carriers?

Covid-19 has various degrees of severity ranging from mild to severe with acute respiratory distress (ARD) and pneumonitis. If there are severe degrees of the infection, there must exist asymptomatic carrier stage. There have been numerous reports confirming the transmission of COVID-19 through asymptomatic carriers via person-to-person contact. Based on the current data, we do not know whether these patients are only asymptomatic initially after contracting the disease or if they are asymptomatic throughout the course of the disease.

Is fever always present in Covid-19 infections?

In a report analysing the clinical characteristics of 7736 hospitalised patients with Covid-19 in China, data regarding clinical symptoms and outcomes in 1099 patients were obtained. Although 88.7% of patients developed fever during their hospitalisations, only 43.8% had fever on admission. Only 67.8% of patients had a cough. These figures refer to only patients admitted to hospital. There are lots more patients with less severe symptoms than that in patients not admitted to hospital.

If we stick to the strict criteria as to who to swab, we are likely going to miss many in general practice that will perpetuate the spread of virus including spreading to health care workers.

I don’t know about you, I can’t tell who amongst the patients that present with flu symptoms have Covid-19 and who haven’t. Thus, I can’t tell whom I should swab and who I should not.

Are radiological abnormalities helpful?

Based on the findings of 1099 cases of COVID, only 14.7% of patients had an abnormal plain chest Xrays. In contrast, 76.4% patients had abnormal chest CT images, in which ground-glass opacity was the most common abnormality (65.5%), followed by local patchy shadowing (48.7%), and interstitial abnormalities (17.0%). You would expect the patients with inflammatory pneumonitis to have pretty obvious ground glass appearance but that’s not always present on plain CXR.

Risk groups – who do we need watch out for?

Hypertension was the most common underlying disease, followed by diabetes mellitus and COPD. However, only 23.7% of patients had a chronic illness which means at least three quarters of patients are “well”. An emphasis on the higher mortality in the elderly is giving doctors and patients the impression that younger patients are spared. This is also leading to less swabbing of younger patients with symptoms. Younger patients with mild symptoms are just told to go home and self isolate for two weeks

Patients with pneumonia were older, with a higher prevalence of smoking history, more underlying diseases, and were more likely to have fever, myalgia/fatigue, dyspnoea, headache, and nausea/vomiting compared to patients with just acute respiratory disease (ARD). In addition, pneumonia cases presented a higher white blood cell count and neutrophil count but had a reduced lymphocyte count compared to ARD cases. Patients with pneumonia were more likely to require oxygenation therapy, mechanical ventilator, renal replacement, and extracorporeal membrane oxygenation, and received more antibiotics and antiviral therapy than patients with ARD. Pneumonia was associated with a higher mortality rate than ARD.

2 weeks of self isolation – whom does that apply to?

The “two week self-isolation” advice is bandied around so frequently that it almost becomes the standard treatment for those with or without Covid-19. We better dissect out the advice.

Scenario 1

If you come into contact with a known case of covid-19 infection or have been on a cruise or have returned from overseas, you are usually told to self-isolate for 2 weeks. While the incubation period is usually 4-5 days in the majority of cases, the incubation period can be as long as 2 weeks. Hence, the period of self-isolation of 2 weeks in these cases. However, that advice is only applicable if the patient has no symptoms.

If the patient has any symptoms of a flu, then he/she needs to be swabbed. If the swab is positive then this patient has Covid-19 and needs to be treated as such. If the patient tests negative, he/she can still be in the incubation period and still needs to serve out the rest of the two weeks of self-isolation.

Scenario 2

Because of the restrictions in who we should be swabbing, we often come across patients with flu symptoms but do not satisfy the criteria to be swabbed. It is not uncommon to send these patients home to self-isolate for two weeks. Is that the correct recommendation?

If this patient does not have Covid-19 infection, I am not sure what the two weeks will achieve. If you don’t swab the patient to confirm or exclude Covid-19, then 2 weeks of self-isolation won’t help us one way or the other. On the other hand, if this patient has Covid-19 infection (and we don’t know that because we don’t swab), then is two weeks of self-isolation enough. The correct number of weeks to isolate is dependent on two negative swabs. This could be take 3-4 weeks. After 2 weeks of self-isolation, the patient might still be contagious. How would you know unless you swab?

What should we do?

I know we need to rationalise our swabbing because we are short of testing kits. Tell us straight and come up with a rational approach and not make us rely on guess work. It makes us complicit in spreading the virus. It also facilitates spreading the virus to health professional colleagues.


Guan W, Ni Z, Hu Y et al. Clinical Characteristics of Coronavirus Disease 2019 in China. NEJM February 28, 2020 DOI: 10.1056/NEJMoa2002032