Obesity’s Role in COVID-19 Deaths: ‘Profound’ and Independent of CVD

An attempt to ‘disentangle’ body mass index from other comorbidities, risks factors, and social determinants of health in the context of COVID-19 has shown that obesity plays a unique and “profound role” in the risk of dying, distinct from other risk factors, particularly for men and younger patients.

While case series and smaller retrospective analyses linking obesity to disease severity in COVID-19 have exploded in the literature over the last few months, this particular analysis—published this week in the Annals of Internal Medicine—is large and includes patients with a positive diagnosis outside of the hospital setting. It also covers a diverse population of patients.

“Our finding that severe obesity, particularly among younger patients, eclipses the mortality risk posed by other obesity-related conditions, such as history of myocardial infarction, diabetes, hypertension, or hyperlipidemia, suggests a significant pathophysiologic link between excess adiposity and severe COVID-19 illness,” write Sara Tartof, PhD (Kaiser Permanente Southern California, Pasadena), and colleagues. “Our findings . . . reveal the distressing collision of two pandemics: COVID-19 and obesity. As COVID-19 continues to spread unabated, we must focus our immediate efforts on containing the crisis at hand. Yet, our findings also underscore the need for future collective efforts to combat the equally devastating, and potentially synergistic, force of the obesity epidemic.”

Overlapping Epidemics

The analysis drew on electronic health records across nine counties in southern California within the Kaiser Permanente Southern California (KPSC) health system. A strength of this particular analysis, stressed Tartof, is the completeness of the KPSC data across an ethnically and economically diverse range of neighborhoods.

In all, 6,919 patients had a positive COVID-19 diagnosis (> 80% confirmed by PCR test) across the nine counties in the KPSC network: 55% were female and 54% were Hispanic; mean BMI was 30.6 kg/m². Comorbidities were common and included hyperlipidemia (23%), diabetes (20%), and hypertension (24%). Overall, 3% of patients died within 3 weeks of their diagnosis, of whom 67% were admitted to the hospital. Among the 97% of patients with a positive diagnosis who survived, 15% required hospitalization.

As Tartof emphasized to TCTMD, the KPSC electronic health records allowed researchers to weigh the impact of patient BMI against a wide variety of individual patient-level covariates, including 20 comorbidities and a range of neighborhood-level factors. Other analyses, she said, have not be able to do. After adjusting for all of these factors, investigators found that high BMI was “strongly associated” with a risk of death, rising in parallel with obesity levels such that those in the highest category (BMI > 45 kg/m²) had a more than fourfold higher risk of dying. Moreover, male patients were significantly more likely to die than women, overall, and this was particularly marked in men with higher BMIs under the age of 60.

“Many cardiac conditions are associated with obesity, so in a given patient you may not be able to disentangle them, but certainly a lot of research has pointed to prior MI and other things as being really important risk factors,” Tartof told TCTMD. “But in a lot of those early papers, BMI wasn't even included as a covariate. So a lot of the risk attributed to cardiovascular diseases I think we need to put in context and also consider BMI as actually a really primary driver of COVID-19 severe outcomes.”

Of note, Tartof and colleagues did not see an increased risk of death associated with race/ethnicity, or with sociodemographic characteristics. “The statistics coming out of the CDC, those are true. There really are health disparities with COVID-19 by race, ethnicity—don’t get me wrong,” she said. In the Annals analysis, Tartof continued, the lack of a signal by race, ethnicity, or social determinants of health likely stems in part from the extent to which they were able to model for other comorbidities, but also reflects the more equitable access to care within the KPSC system, underscoring the importance of healthcare access in equalizing outcomes.

What Can Be Done?

In an accompanying editorial David Kass, MD (Johns Hopkins University School of Medicine, Baltimore, MD), estimates that there have been nearly 300 articles to date documenting a “dose-dependent” link between severe obesity and increased morbidity and mortality in COVID-19. This latest paper, writes Kass, on top of the consistent findings from earlier work, should “put to rest” the notion that obesity is only common in severe COVID-19 because it is common in the population. Instead, he writes, obesity needs to be considered as an important and independent risk factor for disease severity.

Kass also floats the theory that the higher risk seen with severe obesity in younger adults likely relates to the fact that other serious comorbidities—most notably cardiovascular—tend to evolve later in life. Moreover, “that males are particularly affected may reflect their greater visceral adiposity over females, given that this fat is notably proinflammatory and contributes to metabolic and vascular disease”—a theory also explored by Tartof et al in their paper.

Even when COVID-19—hopefully—goes away, we'll still be left with the obesity epidemic and hopefully we can start to address it a little more effectively. Sara Tartof

To TCTMD, Kass stressed that the pattern emerging in the COVID-19 literature should be easily recognizable to cardiologists. “These are all the same comorbidities,” he said. “Everything that cardiologists are battling in patients with cardiovascular disease, where obesity at this level starts to really impact things adversely and is recognized as a risk, are precisely the same underlying pathobiological components that have turned obesity into a COVID risk. It's not unique to COVID-19: what obesity does to the immune system, to the metabolism, to vascular health, to the heart and heart structure and heart remodeling—these are all things that having a BMI of 40 does. It doesn't kill you, but it definitely changes your body.”

Both Kass and Tartof acknowledged that there’s no easy fix, and said that physicians encountering a COVID-19-positive patient who also has a high BMI need to be aware that this is a patient in whom severe illness and death are more likely.

“Knowing early on, even at the point of testing, if someone is in one of these extremely high-risk categories, I think that the follow-up care for those patients should probably be a little more intense,” Tartof said.

Kass agreed, making the additional point that the risks associated with obesity need to be a part of the “messaging” to the public, so that the people most at risk take precautions to avoid getting infected in the first place. Many of the rural, more economically depressed regions now being hit hard by COVID-19 are also those with some of the highest rates of obesity, he noted, yet these are some of the same populations for whom advice for preventing infection such as social distancing and wearing a mask are falling on deaf ears. “Unfortunately we’ve turned this into a political thing instead of a medical thing,” said Kass.

Down the road, Tartof hopes that the lessons of COVID-19 will refocus attention on the soaring rates of obesity. “The cardiovascular disease community would already know this, that the obesity epidemic underlies a lot of other health challenges,” Tartof said. “Even when COVID-19—hopefully—goes away, we'll still be left with the obesity epidemic and hopefully we can start to address it a little more effectively.”