Studying COVID-19 and obesity

On March 8, David Leonhardt wrote in the New York Times about the “mystery” of low COVID mortality in many low-income, resourceconstrained countries in Africa and Asia. Framed differently, the mystery in question becomes: why is the COVID mortality rate so high in high-income, resource-rich countries like the United States? The answer likely hides in plain sight.

A contribution of obesity to COVID risk is a recurrent theme in research on the topic, and thus — epidemiologically consistent. The

CDC just reported a clear dose-response relationship between BMI and COVID risk, a finding supportive of causal influence. There are, as well, clearly plausible pathways for the putative effects.

The mechanisms of an obesity influence on COVID severity are likely both metabolic (e.g., inflammation, impaired immune responses, compromised vascular health, endocrine imbalance, etc.) and mechanical (i.e., severe obesity compresses the thorax, reduces respiratory tidal volume, makes repositioning difficult).

When I volunteered on the front lines in NYC during the surge, I saw that latter effect very clearly; there was an immediate benefit in oxygenation with repositioning, and with severe obesity, it was very hard, for patient and staff alike, to achieve that on an emergency department gurney.

There has long been reference to a particular “obesity paradox” that has largely been debunked — the notion that obesity at times confers a health advantage. Other things being equal, it does not. But there is another, global, obesity paradox that has much clearer validity: obesity tracks with the opportunities of affluence in relatively indigent populations, but tracks with the burdens of indigence in affluent societies.

Obesity and bad COVID outcomes might be “correlation without causation,” if something else associated with both, lack of access to good medical care, for instance, is the actual causal agent. Such agents in epidemiology are known as “confounders.”

In relatively indigent countries, perhaps obesity signifies lower risk for individuals, because the individuals vulnerable to obesity tend to have social and economic advantages relative to the population at large. So, obesity might correlate with better access to good medical care, less crowding, and so on.

Here in the United States, obesity is most severe where social, economic, and environmental disadvantages

(not relative lack of will-power or personal responsibility) make it so. Unbundling the specific contributions of these various factors to COVID risks and losses is nuanced and subject to on-going effort. When that effort is done, I predict that the charges against obesity will persist.

The weight of evidence is already more than ample to indict, if not convict, the prior pandemics of chronic, cardiometabolic disease and obesity as accomplices to the grim assaults of COVID-19. Moreover, in the United States, where the over 500,000 losses to COVID we have suffered rightly garner collective, reverential sorrow — these antecedent pandemics claim more losses to premature death among us every year. Poor diet quality alone, a parameter over which we exercise control, causes 500,000 or more premature American deaths annually. Yet these “prior pandemics” that killed hundreds of thousands of us yearly on their own, then aided and abetted SARSCoV-2 in killing hundreds of thousands more of us over the past year — garner no respect, no collective mourning, and arguably, little notice at all. They have been consigned to the background noise of epidemiology.

Why? Perhaps, in part, because these are generally slow-motion perils that elude the fast-action potentials of the human fight or flight response. Perhaps, in part, because familiarity breeds contempt, even for the causes of mass casualties. Or, perhaps because the propagation of obesity and chronic disease is good business in America; costly for the many, but highly lucrative for few. Perhaps the massive marketing might of relative few has effectively whispered us into the trance that consigns this to the acceptable background noise of modern epidemiology