Why the Covid-19 lab-leak theory is a big deal

Spoiler alert: it has nothing to do with proving the origin of Covid-19, pointing the finger, apportioning blame, reparations, or US-China relations.

The lab leak theory

The lab leak theory, as is now widely understood, is the hypothesis that Covid-19 invaded the human population, not as a result of some ‘natural’ animal-human interaction, but as a result of a coronavirus making its way from laboratory samples directly into the world. The most likely place this might have occurred is at the Wuhan Institute of Virology.

To be clear, the lab-leak theory does not rest on there being malicious intent, or gain-of-function research, or negligence. It is enough that a dangerous pathogen might have accidentally emerged from a high security laboratory to wreak havoc on the world. The lab-leak hypothesis does not depend on human engineering of a virus. A naturally occurring (though no less dangerous) virus could still leak from a lab.

A friend of mine expressed the feeling many people have about Covid-19 origins when she said: “I don’t think it really matters. Pandemics will arise from a variety of sources, what’s more important to focus on if how can we learn (from this experience) how to respond better when they do?”

This is part of the crucial point. But our best response to a pandemic is to prevent them. If we are dismissive of a laboratory origin, then our prevention strategies will overlook prevention at laboratory sites. If the laboratory hypothesis is reasonably plausible, then we ought to invest a reasonable amount in securing laboratories. If we are ever able to elucidate the actual origins of Covid-19 and it was in fact a sequence of events involving a laboratory, then our investment in prevention should specifically target that sequence of hazards.

I personally doubt that we will ever find conclusive evidence of a lab leak. This is partly because the leak may have been a leak of a naturally occurring organism, and partly because any hard evidence in supporting a lab leak has almost certainly been destroyed. We should keep investigating, just in case, but I’m not holding my breath.

The probability of a lab-leak

Without conclusive evidence we must take a probabilistic approach. Authors of one account derived the probability of the Covid-19 pandemic being the result of a lab-leak by imputing known published probabilities across a range of key causative variables. On this principled probabilistic account, the probability of a naturally occurring SARS-like community outbreak originating in Wuhan is 1 in 1000 years, while the probability of a lab-leaked community outbreak occurring in Wuhan is 1 in 833 years. Given that there was in fact a SARS-like community outbreak in Wuhan, one of these must have occurred. This means the probability of a lab-leak origin is 55% and a natural origin is 45%. Even under very conservative assumptions these authors still find 6% probability for the lab leak hypothesis. This is the correct approach to reporting on the origins of Covid-19.

We shouldn’t be pitting one theory against the other to determine certainty, which we may never find, we should be determining probability, which can then guide resource allocation.

Hence the reason why the lab-leak theory is a big deal, and it is a very big deal, is simply because it is plausible, and not only plausible, but perhaps equally likely.

Irresponsible representatives and media

There is almost no greater threat to humanity than a global pandemic. This means that we must avoid them at all costs. Unfortunately, getting to the bottom of the issue has become a partisan football. This is clear when former US President Trump spoke of the ‘China virus’, or through the combative imagery, tone and rhetoric on Sky News Australia when that channel reports on the lab-leak hypothesis.

Much media coverage of the Covid-19 origin story is not helping us prevent future pandemics. In fact, it is probably raising the risk. Too many media outlets are running an attack campaign on China. The effect of this kind of approach is likely to push Chinese officials into defensive mode, counter-offensive mode, and secretive mode. None of this helps us achieve a transparent investigation of the facts. This is a lose-lose strategy, and media outlets running this approach will have the blood of future pandemics on their hands.

Other media outlets are running the opposite campaign. They are minimising the lab-leak theory with faulty logic such as ‘there’s no conclusive evidence for a lab-leak’ (when there’s no conclusive evidence for a natural origin either), or ‘Leading biologist dampens his “smoking gun” Covid lab leak theory.’ Which misses the point that a lab-leak of a naturally occurring organism might not carry a ‘smoking gun’ genetic signature. None of this is helpful.

High-reliability industries and system safety

Many industries depend on ultra high-reliability functioning. These are industries where one failure can have dire consequences. The rate of error needs to be as low as one in a million, or even less. We don’t want any more than one in a million patients to die due to anaesthesia, or one in a million planes to crash, or one in a million nuclear control rods to jam. Ideally the failure rate is even less. In surgery, these are the ‘never events’.

These high-reliability industries, provide a template for the way to investigate causal factors when disaster (or near miss) happens. The process must be a robust, dispassionate, no blame approach. People are error-prone, this is simply a fact of our psychology and biology. Therefore, systems must be engineered to make failure impossible. It is inadequate systems that cause catastrophe because the systems have not adequately protected against error-prone humans.

In laboratory science we never want a dangerous highly communicable organism to infect a human. In fact, the downside is so great (potentially millions if not billions of lives at risk) that we should aim to have complete knowledge of any situations where a laboratory safety failure might possibly have occurred, or is predisposed to occur, or has occurred with 50% probability!

If the lab-leak theory is equally likely to the natural origin theory, then we should assume that there was a lab-leak and spend a decent amount of resource conducting a no-fault investigation into how-possibly it might happen. What causal factors could lead to a lab-leak. This kind of process should draw upon the well-established principles of root cause analysis and failure modes and effects analysis from engineering, as well as human factors principles deployed in high-reliability systems. If any flaw in the laboratory system is identified, then this must be addressed, and the learnings shared with all other labs and also with regulators. All this should occur whether or not we ever find conclusive proof a lab-leak happened.

There should be no witch hunt, there should be no talk of fault and blame and reparations. In fact, there should be a guarantee that no fault will be determined. We want cooperation, transparency and access, but sadly the approach to date has been combative and finger pointing. Representatives’ comments fuel headlines and headlines fuel the agenda of representatives in a vicious cycle. Each actor in this system whether editor or politician has a responsibility for nurturing the conversation in a direction more likely to save future lives than cost them.

Greatest threat to humanity

Covid-19 is, thankfully, a serious but not catastrophic pandemic. Next time we might not be so lucky. An engineered virus, possibly the result of gain-of-function research, could escape a laboratory. It could cause another global pandemic, it could kill a billion people. All this makes the threat from bioengineering a greater, or at least more immediate risk than climate change, a risk equivalent to nuclear war, a risk potentially more imminent than a catastrophic failure of advanced artificial intelligence. Seen in this light the world ought to be far more interested in preventing the next pandemic, far more fastidious about identifying all ways that a virus ‘how-possibly’ might have escaped the Wuhan Institute of virology, and spending far less time and effort waiting for conclusive proof, and seeking to blame, effort that is preventing us preventing the next pandemic.

Even if the probability of a lab-leak at Wuhan is ‘only 6%’, but especially if it is 55%, then decision makers need to pay attention. There is no point waiting for a ‘smoking gun’. Biosafety and biosecurity should be ultra high priority policy domains moving forward.

There is in fact a blueprint for how to begin the process of preventing lab-leaks. The Global Health Security Index (GHSI) prescribes 140 things that countries can do to enhance health security. Many of these items pertain to biosafety and laboratory security, and almost no countries have implemented any of them to date. It is unfortunate that some measures of Covid-19 ‘success’ have not correlated with the GHSI, and hence some researchers have criticised the GHSI’s usefulness. If lab-leak is a live hypothesis, this mindset might mean throwing the baby out with the bathwater.

The Climate Change Commission in New Zealand has recently recommended that changes required to cut emissions to acceptable levels might cost 1.2% of GDP by 2050. At a global level this is US$1 trillion dollars per year. Appropriate investment in biosafety and biosecurity, including contributions to new and needed international organisations probably does not need come anywhere close to this, but such investments are an equal priority given the risks and probabilities involved. It just so happens that US$1 trillion dollars per year is the expected average annual cost of emerging infectious diseases.

Some first steps are obvious, and free. The world banned atmospheric testing of nuclear weapons, and we can ban gain of function research in virology.