Testing times

Since the start of the COVID-19 pandemic, scientists have been moving at pace to develop tests to help stop the spread, assess the extent and understand more about COVID-19. There are currently two types of test available – Polymerase chain reaction (PCR) and antibody tests. PCR testing is used to confirm if a person currently has the virus. Serology testing aims to detect if they’ve had the virus previously and as a result have developed antibodies.

Whenever any test is evaluated for approval several things are considered: 

• Accuracy – how sensitive or specific a test is.
• Reliability - how reliable a test is in different settings – does it produce the results you’d expect in a consistent way?

It's clinical insight like this, together with factors such as safety, that we use at Aviva when we’re reviewing the appropriateness of new tests for customers. We then apply the principles of private medical insurance to assess whether it should be included in our cover.

Benefit may vary depending on the cover the individual has selected. However, in general, we’ll consider cover for tests which inform a diagnosis or positive action associated with treatment covered by an individual’s private medical insurance policy. For example, we’ll cover investigations to help a clinician diagnose diabetes but routine monitoring following that diagnosis, is typically excluded from cover.

Similarly, if a customer’s doctor has a high suspicion that they have prostate cancer, we’ll pay for the individual to have a prostate-specific antigen (PSA) test as part of the diagnosis process. However, we wouldn’t cover a PSA test as a screening tool if the individual didn’t have any symptoms of cancer. This is because PSA is a good test to detect and monitor prostate cancer but isn’t reliable as a screening tool. An individual’s PSA levels can be raised for many reasons and in some cases, the test can suggest prostate cancer when no cancer exists (a false-positive result). The PSA test is very sensitive but not specific enough to be meaningful.

We also assess clinical appropriateness and safety when considering funding for tests which are used post diagnosis. If a test helps identify the most appropriate treatment option for a customer which could positively impact their experience, we’ll consider it for cover under the terms and conditions of their policy. This includes tests such as molecular profiling to help an oncologist’s clinical decision-making process.

A focus on COVID-19 testing

In the case of COVID-19, we’ll pay for PCR and antibody testing as part of a diagnosis. If a customer has symptoms which their specialist feels could be caused by them either having, or have had, COVID-19, we’ll pay for them to have a test to confirm if this is the case. A typical example could be a customer experiencing severe fatigue which is a known symptom of long-COVID but could also be caused by several other conditions. 

We’ll also pay for customers who are about to undergo private medical treatment to have a PCR test to identify whether they have the virus. A study published in the Lancet earlier this year ^{[1]Footnote 1}, suggests that there is a material increase in risk in going through elective surgery if an individual has the virus. By funding the test, we’re able to help our customer’s doctor decide whether it’s safe to continue with their treatment. Testing pre-surgery, also helps to protect the medical team and control the risk of infection in the hospital. 

However, we won’t cover PCR testing or antibody testing for asymptomatic customers who aren’t about to undergo private medical treatment. While customers may find it useful to know if they currently have or have had COVID-19 in the past, we don’t cover the test as a screening tool. There are a number of reasons for this, and a key one relates to reliability. 

In order to meet the Medicines and Healthcare Products Regulatory Agency (MHRA) guidelines, COVID-19 tests need to deliver more than 98% specificity and sensitivity. This means that the results are deemed accurate and they perform in a consistent way, in consistent settings in nearly all cases. While the COVID-19 PCR test appears to be delivering consistent and statistically viable results, there are still challenges with the antibody test. This is in part driven by the fact that the virus’s behaviour isn’t yet fully understood. 

While the majority of test results should be accurate, there will be some people who record false-positives and false-negatives. This means that people may be told that they don’t have antibodies when they’ve had the virus. And, there will be those who appear to have antibodies when they were asymptomatic. 

Timing also plays a key role in the accuracy of tests. For example, current evidence suggests that if the test is done within 20 days of the onset of symptoms, there’s a risk of a false-negative if the body hasn’t had enough time to generate the antibodies. 

Accuracy of results is crucial when considering what actionable insight an individual would take from the results, as there is still a lot of uncertainty about whether having the antibodies provides any immunity. For example, how would the knowledge of the presence of antibodies impact an individual’s behaviour and the behaviour of others towards them? 

Would they feel invincible – and no longer ‘stay alert’ possibly putting others at risk? Could a positive result lead to other social issues such as inequality, for example, by being encouraged back to work along with others who have tested positive, or being allowed more access to travel and ‘everyday life’ than those who don’t have antibodies? And, would those who have tested negative feel anxious that they’re still at risk? 

Given that there’s still so much to learn about the virus, the assumption that a positive antibody test automatically means immunity is a dangerous one. With any type of medical science, there are always exceptions to the rule. 

Without enough evidence to link the presence of antibodies to immunity, testing might cause more harm by falsely reassuring people who test positive for antibodies. Moreover, even when antibodies are detected, we don’t know how long they’ll exist in the body. This raises the question of whether a repeat test may be required in the future. 

The concern that testing could lead to inequality is a valid one. While, the WHO’s briefing paper on “Immunity passports” ^{[2]Footnote 2} sounds interesting, if this approach were to be adopted, it needs to be introduced with caution. There’s a risk that it could become another factor of inequality if it’s socialised without working out all the unintended consequences. 

Over the past months we’ve seen exceptional developments in medical science. Tests and vaccinations are being developed at unprecedented speed. And, every day we learn more about the COVID-19 virus. There’s still a way to go but we’re seeing positive steps. Guided by the expertise of data scientists and clinicians, we’ll learn to live within our ‘new world’, hopefully bringing with us some of the positive behavioural and societal changes we’ve adopted during this time.