Edge Collective

34. The real risks of Long COVID

A review of recent research on the rates and impacts of Long COVID.

Suggested readings:

"Long COVID is harming too many kids" -- Blake Murdoch, Scientific American 11/18/2024.

"Long COVID in children and adolescents" -- Lopez-Leon et al, Nature Scientific Reports 6/23/2022

"Long COVID: major findings, mechanisms, and recommendations" -- Davis, Topol et al, Nature Reviews Microbiology 1/13/2023

Higher rates and greater impact of long COVID than previously thought

Most people now seem to believe that COVID-19 has become a 'mild' disease, and that -- aside from precautionary vaccination against the odd chance of a more severe case -- it is no longer worth taking many (or any) precautions against infection.

Indeed, while some people still do feel 'knocked out' for a few weeks after infection, most COVID-19 cases now seem to be 'mild', especially for those with up-to-date vaccinations.

There is, however, a growing body of research indicating that COVID-19 causes long-term suffering in many people -- a condition referred to as 'long-covid' -- even after initially 'mild' infections; estimates vary, but there seems to be consensus that at least 7% and perhaps as many as 25% (1 in 4) of COVID-19 cases result in long-term symptoms, including brain fog, severe fatigue, and an increased risk of Type I diabetes. For some, these longer-term symptoms resolve within a few weeks, or months; for others, they have not yet resolved.

Further: it has been shown that the risk of developing 'long covid' symptoms increases with every infection, whether initially mild or severe. By the third infection, the chances of developing some long-covid symptom is estimated to be over 35%, or greater than 1-in-3.

)
Chart 2 from Kuang et al, "Experiences of Canadians with long-term symptoms following COVID-19", Statistics Canada 12/8/2023.

It is now understood that COVID-19 can impact every organ system in the body, including the heart and the brain. Even mild COVID-19 infection is now known to significantly increase the chances of severe heart disease, stroke, and dementia.

)
Figure 1 from Davis, Topol et al, "Long COVID: major findings, mechanisms and recommendations", Nature Reviews Microbiology 1/13/2023
)
Figure 2 from Davis, Topol et al, "Long COVID: major findings, mechanisms and recommendations", Nature Reviews Microbiology 1/13/2023. 'Hazard ratios' represent the relative risk for a person with a COVID infection relative to someone not infected. For comparison, the hazard ratios for smoking 10 to 20 cigarettes a day for stroke, cardiac arrest, diabetes, pulmonary embolism, and death are 1.61, 1.5 to 2.5, 1.3 to 1.5, 1.2 to 1.5, and 1.8, respectively.

In fact, researchers no longer characterize COVID-19 as a 'respiratory-only' disease, but as 'vascular' disease. One theorized mechanism of COVID-19 damage, for which there strong recent supporting experimental evidence (see 2024 research concerning the ability of the COVID-19 virus to induce clotting, [here]((https://www.nature.com/articles/s41586-024-07873-4), with explanatory review here) involves the formation of 'micro-clots' in blood vessels; these microclots can then disrupt normal functioning in all organ systems.

There are currently no well-established effective treatments or cures for long covid; for some, it appears to be a life-long condition, and is in many cases debilitating.

Impact of Long COVID on children

The latest estimates suggest that between 10 to 20 percent of children infected with COVID go on to experience long-term symptoms, including:

A recent JAMA study (Gross et al, "Characterizing Long COVID in Children and Adolescents", JAMA, 8/21/2024) noted that long COVID symptoms in children "affected almost every organ system".

Figure 2 from Lopez-Leon et al, "Long-COVID in children and adolescents: a systematic review and meta-analyses", Nature Scientific Reports 6/23/2042

Studies indicate that COVID infections can result in long-term damage to

as well as increasing the relative risk of new onset pediatric diabetes (see "SARS-CoV-2 Infection and New-Onset Type 2 Diabetes Among Pediatric Patients, 2020 to 2022", Miller et al, 10/14/2024)-- even in children with mild or asymptomatic infections.

Why are more people not taking precautions?

Despite the relatively high risk of this potentially debilitating condition, few seem to be taking precautions against infection.

One reason may be that long covid research does not currently seem to be a significant part of mainstream discourse; few seem to be aware of the results of recent studies.

Another reason may be the delay between initial (often mild) infection and the later emergence long-covid condition, which makes it difficult for most people to make the connection between their ongoing symptoms and COVID-19.

“Not all people even know that their condition might be caused or exacerbated by COVID,” Estiri said. “So those who go and get a diagnosis represent a small proportion of the population.”
from "Long COVID may be far more common than previously known", Boston Globe, 11/16/24"

Yet another reason that few people seem to try to avoid infection currently is that the risks of longer-term COVID-19 impacts seems generally to have been minimized.

Understandably, the initial focus in the COVID-19 pandemic was on preventing or reducing severe illness or death due to the virus; and initially, no long-term data was yet available. But even some of the longer-term research into COVID's impacts has ended up providing a false sense of security around long COVID.

For example: a study in a prominent medical journal reported in 2023 that the rate of long COVID symptoms in children was 'strikingly low', at approximately 0.4 percent, a result that was widely publicized. Subsequently, this study was retracted due to methodological errors; more recent studies now estimate the rate of long COVID in children to be between 10 and 20 percent. (See "Long COVID is harming too many kids", Blake Murdoch, Scientific American 11/18/2024.)

Weighing the costs of risk reduction

A final reason that so little attention currently seems to be paid to long covid risks is that for many, the actions required to significantly reduce the risk of infection are either impossible, or would incur too great a financial or social burden.

Masking when in shared indoor spaces can often be highly incovenient, or socially charged. Eating in restaurants poses a high infection risk, but is a cherished activtiy that many would refuse to give up.

Even the use of indoor air purifiers is often now seen as being 'overkill' for a disease that most believe no longer poses significant risks.

What could be done?

If people fully understood the risks of long COVID and were to take them seriously, there are relatively low-cost measures that might significantly reduce infection risk, including:

More signifcant and costly changes might include the development of forms of life that allow for social and work interactions to occur within a relatively 'protected' pod of people, which then interacts with the rest of the world in either high-ventilation settings (e.g. outdoors) or while masked. Such forms of life are not so unusual, historically: many rural and even suburban communities allow for this sort of protective 'isolation' to be maintained.

There is also, of course, the hope that an inexpensive and widely-avaiable vaccine that prevents infection might become available. It is hard to predict when or whether such a vaccine would become available.