Although COVID-19’s designation as a national and public health emergency is nearing its end in the US, post–COVID-19 condition, also known as long COVID, seems to be sticking around. Now, new research adds to evidence that long COVID may be affecting the workforce. A survey-based study of more than 15 000 people with prior COVID-19, published in JAMA Network Open, found that those with self-reported long COVID symptoms were more likely to be unemployed.
The employment research builds on an October study focused on the prevalence of long COVID. Both analyses used the same US data set derived from an internet survey called the COVID States Project. Despite its name, the survey casts a wide net of questions related to behavior and beliefs, including many items unrelated to COVID-19. Respondents are also asked about their political stances and social networks, for example. This means the survey doesn’t introduce bias about the illness, Roy Perlis, MD, MSc, lead author of both studies, said in an interview with JAMA.
The survey attracts a range of respondents, Perlis said, including those without strong opinions on COVID-19. And because each survey wave was conducted across all 50 states and the District of Columbia, the research team was able to draw broad conclusions about the nation.
While Perlis’ initial study gauged pervasiveness—it found that 15% of respondents with a prior positive COVID-19 test result reported long COVID symptoms—there were other topics he wanted to explore.
“One of the things we weren’t able to get into that I regretted at the time was the functional implication of long COVID,” said Perlis, who is associate chief of research in the Department of Psychiatry at Massachusetts General Hospital and directs the hospital’s Center for Quantitative Health. (Perlis is also an associate editor at JAMA Network Open but was not involved in either article’s editorial review or decision to publish.)
Specifically, he wanted to determine the extent to which long COVID—especially cognitive symptoms—interferes with the ability to work.
The survey responses were collected between February 2021 and July 2022 in 8 waves, each conducted every 4 to 8 weeks. All 15 308 respondents were US residents aged 18 to 69 years who had a positive COVID-19 test result at least 2 months before being surveyed. Those who reported still having symptoms were also asked to complete a checklist of 25 commonly reported symptoms ranging from dizziness to shortness of breath.
The researchers later combined memory impairment and brain fog—defined in the analysis as difficulty concentrating or focusing—into one variable. Prior research hasn’t focused enough on cognitive symptoms that are pertinent to understanding long COVID, Perlis explained. “It’s not to diminish all the other symptoms—the autonomic symptoms that people experience, the fatigue, the respiratory symptoms—but my focus is on the brain aspect,” he said.
Overall, 2236 of the respondents, or almost 15%, reported having long COVID symptoms. About 12% of individuals with long COVID were unemployed compared with nearly 9% of those who didn’t have the condition. And having long COVID was associated with a higher likelihood of being unemployed, as well as a lower likelihood of working full-time, even when the researchers adjusted for sociodemographic factors such as age, sex, region, and race and ethnicity.
As for the cognitive symptoms, brain fog or memory impairment were reported by almost 46% of respondents with long COVID—and were associated with a lower likelihood of working full-time.
There’s a tendency to dismiss these symptoms, Perlis remarked. However, the results suggest that cognitive symptoms are not only “important because they’re distressing to people, but they’re also important because they have real implications in terms of function.”
Danielle Sandsmark, MD, PhD, an assistant professor in the University of Pennsylvania’s Division of Neurocritical Care, said the findings are in line with her clinical experience. Sandsmark helps lead the Penn Neuro COVID Clinic, which treats patients with neurological symptoms stemming from long COVID, but was not involved with the new study.
“In general, I find that cognitive effects and fatigue are the primary reasons that I hear from patients as to why they are unable to return to their jobs,” she wrote in an email to JAMA. “Cognitive symptoms, in particular, are not associated with an outward, physical disability, but these data demonstrate that these symptoms are associated with a real-world effect, like going back to work.”
When Perlis’ team looked at the subset of respondents who were currently unemployed, only about 28% of those without long COVID worked full time before the pandemic compared with nearly 40% of those who reported symptoms.
This finding “reinforces the point that you’re absolutely getting some people who were not disabled before who are disabled now,” Katie Bach, MBA, MSc, a nonresident senior fellow at the Brookings Institution who was not affiliated with the study, said in an interview.
Bach, who researches the economic impacts of long COVID, wrote early last year that about 1 million full-time equivalent workers in the US could be missing from the workforce at any given time because of long COVID. In a report last August she estimated that, on the lower end of the spectrum, long COVID may cost around $170 billion in lost wages—nearly 1% of the total US gross domestic product.
“Acute COVID is a pandemic; long COVID could be a mass disabling event,” Bach and David Cutler, PhD, a professor of economics at Harvard University, warned in an October opinion piece in the Boston Globe.
More recently, the New York State Insurance Fund—the state’s largest workers’ compensation insurer—published an analysis of 3139 established COVID-19 compensation claims filed between January 2020 and March 2022. Almost a third of the claims involved long COVID. The report concluded that approximately 18% of claimants with long COVID hadn’t returned to work for more than a year. Most of these claimants were younger than 60 years.
But according to Perlis, claims data may not adequately capture long COVID trends.
“There are a lot of people struggling with long COVID who aren’t necessarily collecting billing claims or going to see their doctor about their symptoms,” he said.
Among unemployed respondents who shared their prepandemic employment status, almost 58% of those with long COVID reported that they were currently looking for work compared with about 64% of those without the condition, according to Perlis’ new analysis.
“I was surprised by how close those 2 numbers were,” Bach added. The proportion of people with long COVID looking for work, she said, “suggests they are not so disabled that they can’t do any job, so is it a question of finding the right job or the right accommodations? Is it, realistically, they are that disabled but can’t afford not to work?”
Bach plans to explore those questions by next investigating whether unemployed people with long COVID could work under certain circumstances.
There are also signs that, at least for some, the condition is abating. A recent Kaiser Family Foundation analysis of data from the Household Pulse Survey, a collaboration between several US federal agencies, found that the percentage of people who reported persistent long COVID symptoms dropped from 19% in June 2022 to 11% in January 2023.
“I certainly think that what we are seeing clinically is that most people who do have long COVID are getting better over time—not always perfect, but better,” Sandsmark noted.
But as Bach pointed out, “there’s this subset of people who just aren’t recovering,” and the risks of long COVID aren’t adequately communicated to the public.
For now, the extent to which long COVID may cause unemployment remains unclear, Perlis said.
The study he published last autumn found that people who received a COVID-19 vaccination before they became infected had a lower risk of developing long COVID symptoms. But he hopes his new findings inspire more focus on treating—not just preventing—long COVID.
“Fortunately, fewer people are dying and being hospitalized in the US, but it would be premature to assume that the impact of the pandemic is really diminishing—it’s just shifting,” Perlis said. “As much as we all want to move on, a lot of us can’t move on yet, and [we] need to be thinking about interventions for people who are still struggling with long COVID.”
Published Online: February 15, 2023. doi:10.1001/jama.2023.0157
Conflict of Interest Disclosures: Ms Bach reported having a board membership with the PolyBio Research Foundation, which studies, among other things, long COVID. Dr Perlis reported receiving personal fees from Burrage Capital, Circular Genomics, Genomind, Psy Therapeutics, and Takeda. He also reported that the COVID States Project is supported by the National Science Foundation and the National Institute of Mental Health. Dr Sandsmark reported receiving a grant supplement from the National Institute of Neurological Disorders and Stroke to study neurological biomarkers in COVID-19 infection (supplement to U01 NS114140: Clinical Validation of Serum Neurofilament Light as a Biomarker of Traumatic Axonal Injury). No other disclosures were reported.