April 6, 2023 | Two recent studies funded by NIH’s Researching COVID to Enhance Recovery (RECOVER) Initiative reveal differences in long COVID diagnostic experiences among different racial and ethnic groups. The findings provide information about variations in symptoms and access to care. They also offer insight into the role played by the diagnostic code that is used for long COVID.
Minority Groups May Experience Symptoms Differently
Data has revealed that Hispanic and Black people have higher rates of COVID-19 hospitalization and death than white people. What isn’t as clear just yet is whether racial and ethnic differences play a role in long COVID, or post-acute sequelae of SARS-CoV-2 infection (PASC), which involves the continuation of symptoms or the emergence of new ones—such as brain fog, fatigue, or shortness of breath—following recovery from the initial phase of illness.
In a study published in the Journal of General Internal Medicine (DOI: 10.1007/s11606-022-07997-1), researchers took a closer look at potential racial and ethnic differences in PASC symptoms and conditions among COVID-19 patients. Their study utilized electronic health record data from five academic health systems in New York City, including over 60,000 COVID-19 patients and nearly 250,000 patients without COVID-19 between March 2020 and October 2021.
The findings revealed notable differences in symptoms and conditions among hospitalized and non-hospitalized patients, suggesting that minority groups could experience different symptoms than white people and bear a bigger burden of PASC.
For instance, one to six months following a positive test, Black patients needing hospital care had greater odds of receiving a diabetes diagnosis or experiencing headaches than hospitalized white patients. Meanwhile, Hispanic patients with severe disease had greater odds of experiencing headaches or shortness of breath than hospitalized white patients.
The researchers also found patterns among COVID-19 patients with mild to moderate disease. Among those who were not hospitalized, Black patients were more likely to be diagnosed with pulmonary embolism and diabetes, and Hispanic patients had higher rates of headache and chest pain than white patients.
“We find that the incidence of potential PASC symptoms and conditions differed significantly across racial/ethnic groups, and that compared to white patients, patients from racial/ethnic minority groups had significantly higher adjusted odds of being diagnosed with a range of symptoms and conditions affecting multiple organ systems,” the study authors wrote.
Improved Understanding, Real-World Solutions Still Needed
As the medical community searches for ways to address the long-term effects of COVID-19 and research into treatment options for PASC continues, the researchers believe that arriving at a better understanding of the disparities observed in this study is crucial, considering different populations may have different care needs.
The reasons underlying these differences aren’t well understood at the moment, but several potential contributing factors were highlighted in the study. After vaccinations became available, racial/ethnic disparities emerged in rates of COVID-19 immunization, for example, and previous studies have indicated that Black and Hispanic people have a greater chance of experiencing severe acute COVID-19.
Minority groups may also experience differing levels of baseline health, access to care, and risk of exposure due to their household makeup or type of job—all of which could potentially factor into the odds of developing PASC.
Physical therapist Heather Swain, who has treated patients suffering from long COVID, has seen first-hand how job type and attitudes toward work can impact health care utilization or lack thereof. While practicing in Arizona, she said Hispanic patients comprised roughly half of her total caseload.
“Work ethic is very important in Hispanic culture,” she told Diagnostics World. Hispanic patients often work long hours or multiple jobs. As a result, they can have a more difficult time accessing healthcare services. In addition, the language barrier can prevent some from seeking care, added Swain, who was not involved with the Journal of General Internal Medicine research.
Previous studies on PASC have focused on identifying rates of PASC among people with acute COVID-19, explained Sabrina A. Assoumou, an infectious disease expert at Boston Medical Center and the Boston University School of Medicine who was not involved with the research, but little was known about the differences that may exist among different racial/ethnic groups. In that respect, this newly published study builds upon our current understanding of PASC.
Studies like this are helpful in terms of revealing race/ethnicity disparities, she told Diagnostics World, but from a health equity standpoint, “what we really need are solutions to address what is seen,” she added.
Observing Long COVID through Coding
In a separate study published in BMC Medicine (DOI: 10.1186/s12916-023-02737-6), researchers studied long COVID through the lens of U09.9, the ICD-10-CM code for “Post COVID-19 condition, unspecified.” To arrive at their findings, the researchers analyzed over 33,000 patients from 34 healthcare systems who received a long COVID diagnosis between October 2021 and May 2022.
The authors concluded that their research provides insights into diagnosis and treatment patterns, possible disparities regarding who is diagnosed and who receives care, and potential disease subtypes.
One notable discovery, they indicated, is that the patients diagnosed using this code tend to skew in the direction of white female individuals living in areas with low unemployment and poverty rates—a finding they say, “requires further research and urgent remediation.”
When a new code is released, it’s not as if everyone is equally onboard and well-trained to use that new code right from the get-go, explained Emily Pfaff, a research assistant professor at the University of North Carolina, Chapel Hill School of Medicine and one of the study’s co-authors. Instead, there is a progressive uptake. Beyond that, every institution has its own coding practices.
For example, an academic medical center with a large long COVID specialty clinic located in a major urban area with high levels of education and affluence will likely experience a quicker uptake of that new code. “That’s not a universal truth,” she cautioned. “There are plenty of sites that are not academic medical centers that are using the code.” But based on their findings, it did appear that sites located in more affluent areas were more apt to use the code.
An equally important factor, Pfaff added, pertains to healthcare access. Even if the new code is being utilized equally across the board, an individual who is uninsured or cannot get time off work to see their doctor may also go unlabeled. “Clearly, the role of access to providers and the economic means to afford long COVID care should continue to be studied for their role as contributors to disparate care and outcomes, as well as sources of research and algorithmic bias,” she and colleagues noted in the study.
Lag Time Means Many Likely Went Unlabeled
The assignment of diagnosis codes that can be utilized within electronic health records doesn’t keep up all that well with new happenings. “It takes a long time for new codes to be created, and approved, and released,” said Pfaff. In the US, the use of an ICD-10-CM code for long COVID took close to two years into the pandemic, meaning millions of people who could have been assigned this code may have missed out on that label during the delay.
This can impact a patient’s ability to be included in research, Pfaff pointed out. On an individual level, it might mean missing out on a call to participate in a clinical trial, for example. And on a population level, some people miss out on being included in deidentified cohorts that researchers rely upon to make assertions. If many patients are without a label, especially from the early stages of the pandemic, they don’t get included in research studies. “Their data are missing, and that means we might be missing something,” she said.
Scientists who conduct research on conditions and rely heavily on diagnosis codes to identify populations of interest should avoid over-relying on the U09.9 diagnostic code, cautioned Pfaff, which could come at the expense of all the patients who missed out on this diagnosis. The research community will need to work together to find alternative ways of identifying these patients to come up with the most inclusive possible definition of long COVID moving forward.
The clinical definition and understanding of the underlying mechanisms involved remain “in flux” for the time being, and a single diagnostic code may not be sufficient. According to the study authors, “it is likely that subtypes of long COVID exist, and such subtypes may correlate with specific underlying mechanisms that should be targeted by different interventions.”
Pfaff pointed to diabetes to help illustrate an example of a condition where different codes refer to varying forms of the disease, such as type 1 diabetes, type 2 diabetes, or gestational diabetes. One of the biggest reasons the different codes are helpful is because those different subtypes all require very different treatments.
Although she is not suggesting that long COVID is exactly like diabetes, Pfaff said that we are seeing a “constellation of different symptoms,” which means the people suffering from respiratory symptoms may not benefit from the same treatments as those who are suffering from issues that are more neurological in nature, such as brain fog or debilitating fatigue.
Different long COVID codes could help differentiate between those subtypes and lead to more precise patient treatments. Pfaff and colleagues indicated that refining the definitions will make it possible to “appropriately define cohorts for clinical studies, provide more precise treatment and clinical decision support, and accurately estimate long COVID’s incidence and prevalence.”
Some Call for Additional Research
“This new evidence suggests that there may be important differences in how long COVID manifests in different racial and ethnic groups,” Mitchell S.V. Elkind, a professor of neurology and epidemiology at Columbia University and chief clinical science officer for the American Heart Association, said in a news release announcing the publication of these two RECOVER studies.
Although this new research adds to our collective understanding of long COVID, additional research is still needed to more fully understand the underlying mechanisms that may factor into these symptom and care access variations and to determine the role played by diagnostic codes assigned by medical professionals, he noted.
“This research contributes to our understanding of symptom clusters in long COVID that may be differentiated by race, ethnicity, and influenced by social determinants of health,” added Gary H. Gibbons, director of the National Heart, Lung, and Blood Institute. “It also provides vital insights into the utility, as well as the constraints, of the diagnostic code now in use for long COVID.”
Paul Nicolaus is a freelance writer specializing in science, nature, and health. Learn more at www.nicolauswriting.com.