Myths about cases
By Roy Breeds, Heike Brunner, Shayne Krige, Ian McGorian
Throughout the Covid-19 outbreak, we have been bombarded on a daily basis with the number of ‘cases’. This has never happened before. We do not keep running totals on news websites of the number of people testing positive with HIV or tuberculosis or flu. Despite this focus on the ‘case’ numbers, few people have stopped to consider what they mean. Here are a few common myths about ‘cases’.
- Case numbers represent sick people
Traditionally, medical ‘cases’ reflect people who are sick with a disease (symptomatic). SARS-CoV-2 is the virus that can lead to the Covid-19 disease. The test used in this epidemic tests for the presence of genetic material from the SARS-CoV-2 virus. People who are positive do not necessarily have the disease Covid-19.
Initially, most of the people going for tests were already sick. However, in many countries – including South Africa – this is no longer the case and has not been the case for some time. Many frontline workers are tested on a regular basis, negative test results are now required for travelling, attending events etc. Contact tracing results in people who are asymptomatic being tested. Many of them are testing positive for strands of virus in their systems even though they are not sick and will never get sick. In the context of Covid-19 ‘case’ numbers are not actual cases in the traditional medical sense of people sick with a disease, but rather they represent the number of people who tested positive for strands of virus.
Over recent months, the number of people being tested across the world has increased dramatically resulting in ‘case’ numbers soaring.
- Cases are contagious
We currently test for SARS-CoV-2 using the ‘real-time reverse transcription polymerase chain reaction test’ (rt-PCR test). The test does not detect the virus itself, but detects the presence of genetic material of the virus called RNA. It does this by amplifying any RNA which is present in the sample. Think of it as a zoom lens on a camera. The cycle threshold (ct) is the number of times the specimen needs to be amplified in order to be able to detect whether or not virus RNA is present. At high cycle thresholds, the test can detect fragments of the RNA. These fragments of the virus, as opposed to complete strands, are not infectious. The test can even detect pieces of the virus that are left over from a previous infection weeks before.
Experts say that a cycle threshold over 35 is too sensitive to be meaningful. The US Centres for Disease Control has shown that patients who test positive at a cycle threshold of 33 or higher are probably not carrying enough of a viral load to transmit SARS-CoV-2 to others. Testing for Covid-19 follows a protocol set by the manufacturers of the tests. Our research shows that cycle thresholds of 37 to 40 are being used in most countries – way in excess of the levels recommended by epidemiologists and virologists. Some countries have changed the cycle thresholds over time resulting in a more sensitive test between the ‘first’ and ‘second’ waves.
The rt-PCR test was never intended to, and cannot, detect infectiousness. A review conducted by the New York Times of three sets of coronavirus testing data from Massachusetts, Nevada, and New York found that up to 90% of patients who tested positive for the virus were what scientists refer to as ‘cold positives’. Cold positives are distinguished from ‘hot positives’ who are actually infected with an intact virus. ‘Cold positives’ have very, very low viral loads and are not ill, not symptomatic, not going to become symptomatic and, most importantly, they are not able to infect others. The New York Times reported that if the rt-PCR test was run at recommended cycle thresholds, up to 90% of ‘cases’ would have been negative.
The way rt-PCR test protocols work, only a yes/no result is produced. They only tell you if a fragment of virus was found. This could be a piece of ‘dead’ virus that can’t make the subject sick or contagious. Even where intact virus is present, the test results do not currently give an indication of whether there was a sufficient viral load for the subject to be contagious.
- Test Results are 100% Accurate
No medical test is 100% accurate. A ‘false positive’ means that the results say you have a particular condition, but you actually do not, whereas a ‘false negative’ tells you that you do not have a condition but you actually do. The errors in the rt-PCR tests used for SARS-CoV-2 testing are largely a result of human involvement in the testing process and vary from place to place.
Research in the UK showed that the rate of false positives is between 0.8% and 4%. This means that in the UK, the PCR test is at least 96% accurate which might sound quite accurate but is in fact not a very good result – especially given the policy decisions being made on the basis of the test results.
Researchers in the United States say that ‘evidence from external quality assessments and real-world data indicate a high enough false positive rate to make positive results highly unreliable over a broad range of scenarios.’
In Tanzania, the President of the country covertly sent samples from a goat, a sheep, and a pawpaw fruit to a Covid testing lab and they came back positive for Covid. The government of Sweden discovered thousands of false positive tests from kits made in China that have been used in the United States and probably elsewhere in the world too.
Carl Heneghan, professor of evidence-based medicine at Oxford University and Dr. Michael Yeadon, former chief science officer at Pfizer have both explained that the inaccuracies in the test mean that the majority of positives, if not all of the positives currently being reported are false positives. This is a mathematical consequence of the false positive rate being greater than the ‘prevalence rate’. By way of example, at the peak of the outbreak in the United Kingdom in March, 30% of the tests conducted came back positive and the false positive rate was at least 0.8%. This means that in April, the tests were showing 308 out of 1,000 people being positive when in fact only 300 were (300 true positives plus 8 false positives). The false positives in March were therefore insignificant. Now, however, very few people have the virus. Currently in the UK about 1 out of 1,000 (0.1%) people tested are positive. The false positive rate for the test is still 0.8%. So, on 1,000 tests, we now get 8 false positives and only 1 true positive. When you consider that a country like Germany is currently conducting over a million tests a week, you can see why these experts have sounded the alarm. It is simply unscientific to be using the test when the prevalence of the virus in the population is so low.
The false positives also contribute to the narrative that people can be re-infected since people who tested positive when they weren’t, then subsequently do fall ill and test positive a second time. The same situation plays out around the world – as testing increases, so the false positives increase.
- The second wave is here
The phrase ‘second wave’ implies an equivalent experience to the initial outbreak – a second wave like the first. Given the increase in testing and the false positive rates mentioned above, we are certainly seeing increases in positive tests. In some countries, we are seeing far more ‘cases’ now than we saw at the height of the outbreak.
However, we are not seeing commensurate increases in hospitalisations or deaths. Much has been made of the fact that hospitals in some countries are running close to capacity. This is not unusual. In the UK, for example, fear-inducing headlines cited hospitals as being at 80% of capacity which, it turns out, is less than in 2019. In 2018, the US healthcare system was overwhelmed by flu and Italy’s healthcare system collapsed under the strain of flu outbreaks in 2004-2005, 2009-2010 and 2017-2018.
There are currently four coronaviruses that have been in circulation for years and that are responsible for a fair number of common colds each year. SARS-CoV-2 is likely to be a virus like a host of others that we live with. It will become a seasonal virus that flares up as winter approaches. This, coupled with policies that still cast the net of diagnoses wide so as to capture deaths with Covid (people who died of say cancer but coincidentally also tested positive for Covid-19) as well as deaths from Covid-19 (the direct cause of death was Covid-19) also explains the mild increase in Covid deaths that some countries are now experiencing. SARS-CoV-2 will, like the other coronaviruses and like flu, sadly be the straw that breaks the camel’s back when it comes to the old and the infirm. As we learn to live with SARS-CoV-2, we will need to focus on overall mortality. If one does this, one sees that many countries that have reinstated lockdowns have no excess mortality. Covid-19 is simply this season’s virus.
In South Africa, many questions are being raised by the high level of adjustments being made to the daily death numbers. It has become commonplace for deaths in the last 48 hours to be under 20 but ‘adjustments’ to be in the hundreds. Nevertheless, our death and hospitalisation numbers show no sign of an increase in transmission of the virus. Zweli Mkhize was famously outed for recently citing inaccurate ‘case’ data for a supposed increase in the Western Cape. Not only did he grossly overstate the increase, but the whole increase is explained simply by an increase in testing.
It was confirmed subsequently that the percentage of positive tests remained constant at about 10%, meaning the increase in numbers can only result from an increase in tests. The Western Cape has been reporting less than 10 deaths per day for months and the trend remains downward. The field hospital built at the CTICC has been closed for months. Seroprevalence Tests show that the Province has one of the highest Covid antibody levels in the world (around 40% of the population tested, acknowledging that it is a small study). There is no Coronavirus family precedent for a second wave in summer (they occur in winter) and there are no signs whatsoever of a second wave in the Western Cape. If there were an increase like we are seeing in Europe (significantly fewer hospitalisations and deaths than the first wave), South Africa is eminently capable of dealing with it.
It makes little sense to base policy on ‘case’ numbers. Hospitalisation numbers and deaths from Covid-19 are far more reliable statistics. Positive test numbers are useful only in the sense that the percentage of positive cases to number of tests done gives an idea of the reproduction rate of the virus. That is, if the percentage of tests that are positive is declining, it means that there are fewer infections in the population. That rate continues to reduce in most of the countries that are now increasing restrictions on civil liberties.
We must all be wary of a second wave being manufactured by increases in testing that are presented to us as an increase in cases. It is foolhardy to focus our attention on one disease alone instead of the overall impact of policy choices we make. When we do focus on one disease alone, we should certainly not focus on positive tests but rather on death and hospitalisation data. That data should be balanced against the deaths and other health outcomes of the policies we choose to implement. How many people will die as a result of lockdown is the question that should be foremost in our minds. It is time to place our collective attention where it has been missing for months: on getting back to life, delivering healthcare to people of all afflictions, supporting small business entrepreneurs, helping the nation’s children recover from a lost year and prioritising those at-risk youngsters who have fallen through the cracks of South Africa’s already unequal education system during the devastating lockdown.