Weekend Reads: Dogs and Parachutes (but Not at the Same Time)

by Kevin Schofield

This weekend's reads are two scientific research papers, both relatively short.

The first is the latest twist on dogs' well-documented ability to diagnose certain diseases by detecting unique scents emanating from those afflicted with them — scents that are not detectable by humans. In the past, dogs have been trained to detect chronic conditions such as diabetes or Parkinson's disease and potentially to predict an impending acute medical event such as a stroke, a heart attack, or an epileptic seizure. Recently, researchers have had some success in training dogs to detect people infected with COVID — a very different kind of "rapid test" that one could imagine would be very handy in airports and other places where people gather.

This latest research paper builds on the work training dogs to detect active COVID cases, to see if they can be trained to detect "long COVID." The peculiar thing about training dogs to sniff out disease is that we don't know which chemicals they are smelling that allow them to differentiate between healthy and unhealthy individuals. Are they smelling the coronavirus itself or some other chemical change in how the human body reacts to it? This is an even more pertinent question for detecting long COVID, since so little is known about the underlying causes for why some people continue to show symptoms for several months after nominally recovering from the virus. There is recent research suggesting that some coronavirus RNA reservoirs might remain in certain types of cells, but that is not yet well-established science.

The researchers collected sweat samples from individuals suffering from long COVID, as well as "control" samples from others who were not, and attempted to train dogs to distinguish between the two. Amazingly, the dogs had zero false positives on the control samples and correctly identified about half of the samples from people with long COVID. And the researchers believe that the means by which they collected samples (individuals self-collected the samples from their own underarms on swabs, deposited them in freezer bags, and mailed them in to the lab) could introduce quality-control issues that might have led to an increased "false negative" rate.

Beyond the validation that dogs can be used to detect long COVID, this research establishes something else that is important: long COVID is a real thing that can be diagnosed by its chemical imprint. The researchers seem to believe that this is strong evidence that viral RNA, and possibly reservoirs of the virus itself, remain in the body for months while long COVID symptoms persist, though that might be a stretch at this point since they don't know what the dogs are actually detecting. Still, it is a remarkable result.

Screening for SARS-CoV-2 persistence in long COVID patients using sniffer dogs and scents from axillary sweat samples

The second paper is a parody that was published in 2003 in the British Medical Journal, intended to poke fun at the medical research community's insistence on rigorous scientific studies in the face of the acutely obvious. Steadfastly adhering to the standard format for medical research papers, it purports to study whether parachutes are effective in preventing injury and death due to "gravitational challenge." The authors argue that up to now the belief that parachutes are effective is the result of "anecdotal observations" and that the principles of evidence-based medicine require that a randomized trial be conducted to confirm that they are a successful intervention. This is especially true, they argue, because parachutes themselves have been implicated in injury and mortality when they fail to open or when they accidentally open at the wrong time; additionally, it is not entirely unheard of for someone to survive a fall from great height without a parachute.

It should be pointed out that a randomized trial studying the effectiveness of parachutes would require some subjects to be given a real parachute and some to be given a "placebo" instead — a non-functioning object indistinguishable from a real parachute until the wearer attempts to deploy it. Distribution of real and fake parachutes would be at random. All the trial subjects would jump out of a plane, and statistics would be collected as to which ones were injured or died.

Not surprisingly, the authors conducted a search of the existing research literature, and they did not find any randomized controlled trials of parachute use. Nor did they find anyone willing to participate in a randomized controlled experiment — though they issue a call to (broken) arms to those who criticize the use of interventions that lack a complete evidence-based research foundation, requesting that they volunteer for just such a randomized, placebo-controlled study of the effectiveness of parachutes.

I write this knowing that some may read this paper as validating the use of, say, Ivermectin to treat COVID. It does no such thing, though it does suggest that religiously wearing a mask in public is a proper intervention to prevent the spread of COVID. The difference, as with the parachute, is twofold: first, we understand exactly what mechanism causes a parachute to slow one's descent and a mask to impede the spread of the virus — whereas we don't understand how taking Ivermectin would defeat the coronavirus. Second, the risks associated with using a parachute or a mask are miniscule, whereas the risks of taking Ivermectin are substantial and well-documented.

As another point of comparison, the CDC subjects COVID vaccines to randomized, placebo-controlled trials before approving them, using the same reasoning: other than at a theoretical level, we don't understand whether the vaccines will produce an effective immunity, and the risks of adverse reactions could be substantial.

The original parody paper was written as a response to a scientific community that had learned to weaponize heavy-handed process requirements to attack or slow down work that challenged their mainstream orthodoxy. Today we are witnessing a political movement weaponizing scientific process to the same ends: a group of people for whom there will never be enough scientific evidence to support the safety and efficacy of COVID vaccines or the value of wearing a mask nevertheless will happily embrace the latest fad "cure" once they have caught the virus themselves.

Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials

Featured Image: Photo by aswphotos134/Shutterstock.com.