There are increasing numbers of new commercially available testing kits for Covid-19 and they all vary quite substantially. So what type of test do you need? how do you know the accuracy of the test? how do you know if the test actually works? and if it does work, what do the results of the test tell you?
Analysing the analyst
Pathologists and clinical laboratory scientists are all too aware of the above questions and the challenges that these questions pose in their day jobs. Quality control authorities exist to test the testers and to assess whether the clinical laboratories are producing a level of quality in their test results. If a test can go wrong, it will, and most diagnostic tests are prone to scores of potential pitfalls that can result in an erroneous reading.
Result - death sentence?
It is important to know a little recent history and the context of clinical laboratory, or diagnostic testing.
The main testing scenarios are:
a) visit doctor who takes appropriate sample and sends this sample to the lab for you,
b) request lab or diagnostic kit manufacturer to send you the sample collection kit, take own sample and send this to a lab,
c) buy a point of care (POC) kit and do it yourself.
With the advent of the point of care testing (POC) emerging around the time of new pregnancy testing kits such as the Clear Blue™, you no longer had to go to the doctor to get get tested and the door was being opened for a whole range of new POC testing kits. A lot of work had to go into setting the 'sensitivity' or 'cut-off' levels for these new tests. Due to developing biological technologies such as 'enzyme immune assay' (EIA, or sometimes called ELISAs) the new tests could now detect very low levels of target analytes. Furthermore, provided there was an antibody to it, these target analyte test repertoires could include pretty much anything - from glucose to cancer markers to chlamydia.
Beta human chorionic gonadotrophin or beta HCG was a very useful early target analyte as this was the marker for pregnancy. With the traditional tests that included methods such as 'agglutination' where the lab might be trying to assess whether or not the sample was visually clumping or not, it could take up to three months of pregnancy before there were sufficient levels of beta HCG molecule to detect by this method. However, the new EIA tests were so sensitive that they could almost immediately detected a raised level of beta HCG even prior to the first missed period, and this caused different challenges including the potential for 'false positive' results, particularly as many eggs are naturally rejected in the early days. At this level of sensitivity there is also a hugely increased risk of other molecules or interferences that can cause a 'false positive' result.
Doctors were concerned about the new DIY kits, not just because of potential bad results, but due to the impact of the result on the individual, for example, would more young people commit suicide at the shock of learning they had become pregnant? Or perhaps worse (in an age prior to anti-retroviral drugs) if the test said you were HIV positive, had you just received a death sentence?
Diagnostic technology creates greater clinical challenge
Of course, the technology has moved on considerably since the advent of EIAs in the eighties and nineties, and the latest genetic tests mean that specific analytes (the molecules you are trying to identify and / or measure) can be detected with accuracy at practically any low level. However, the fundamental rules around 'sensitivity' (how low can you go?), 'specificity' (how sure are you that you are picking up what you want?), quality control and clinical 'meaning' all remain the same.
As ever, more information raises more questions than answers. For example, you might be told that you have a positive result for Covid-19, but what are you actually positive for?
- Are you positive for the virus itself, or form some form of antibody to the virus?
- How do you know that this result is accurate?
- How do you know that the test used has been properly evaluated?
- Has the result been confirmed by another or second method?
- Do you just accept the result as a 100% fact?
- What does 'being positive' for Covid-19 actually mean?
- Does it mean you have active virus?
- Does it mean you are a carrier?
- Does it mean you are about to get sick?
- Should you be in quarantine? Or,
- does it mean that you are now immune?
It is one thing being tested by an accredited clinical laboratory with an expert to explain the implications of the result, but self-testing with a DIY kit may raise more questions and cause more anxiety than it is perhaps meant to prevent.
Even if the result is accurate and you are positive for the presence of Covid-19, limited information is available to characterise the associated spectrum of clinical illness, there is no vaccine (yet) or specific treatment available (yet) and the level of care is therefore only 'supportive'. With regards to testing, the US Centres for Disease and Infection (CDC) are advising doctors with "interim guidance" to work with state health departments to coordinate testing through public health laboratories. They are recommending only the use Covid-19 diagnostic testing, "authorised by the Food and Drug Administration". Measures have now also been put in place, down as an Emergency Use Authorization (EUA), to make more testing kits available to clinical laboratories. CDC report that, "this additional testing capacity will allow clinicians to consider COVID-19 testing for a wider group of symptomatic patients". In the meantime, CDC are recommending that, "clinicians should use their judgment to determine if a patient has signs and symptoms compatible with COVID-19 and whether the patient should be tested".
Further advise from the CDC says that clinicians will be able to access laboratory tests for diagnosing Covid-19, "through clinical laboratories performing tests authorized by FDA under an Emergency Use Authorization (EUA)".
FDA has not authorized any test that is available to purchase for testing yourself at home for Covid-19
This guidance has caused a little bit of confusion amongst the existing and emerging new commercial manufacturers of kits. At least four new start up diagnostic companies were planning to roll out DIY tests, but the Food and Drug Administration (FDA) released a statement saying: “We want to alert the American public that, at this time, the FDA has not authorized any test that is available to purchase for testing yourself at home for Covid-19.” However, some of the companies insist that DIY coronavirus tests are permitted under the government rules loosened to expand the availability of kits.
in the UK, Prime Minister Boris Johnson has announced that pathology laboratories associated with the NHS will have expanded capacity to provide testing facilities.
See also CDC FAQs on lab testing for Covid-19