A Massachusetts General Hospital (MGH)-led team has developed a preliminary definition of long COVID based on 12 symptoms that affect infected patients more often than uninfected people 6 months or more after a positive SARS-CoV-2 test.
The researchers, who described their work today in JAMA, analyzed observational, prospective data from 9,764 adults surveyed about persistent symptoms and their severity at 85 sites in 33 states, Puerto Rico, and Washington, DC, participating in the Researching COVID to Enhance Recovery (RECOVER) trial by April 10, 2023, arriving at 12 defining symptoms.
Median participant age was 47 years, 71% were women, 16% were Hispanic, and 15% were Black.
The authors noted the lack of consensus on what constitutes long COVID, or postacute sequelae of SARS-CoV-2 infection (PASC). "Most existing PASC studies have focused on individual symptom frequency and have generated widely divergent estimates of prevalence due to their retrospective design and lack of an uninfected comparison group," they wrote.
"Moreover, defining PASC precisely is difficult because it is heterogeneous, composed of conditions with variable and potentially overlapping etiologies (eg, organ injury, viral persistence, immune dysregulation, autoimmunity, and gut dysbiosis)," they added.
23% of infected patients had qualifying long-COVID score
Of 2,231 participants first infected with COVID-19 on or after December 1, 2021, and enrolled in the National Institutes of Health–sponsored study within 30 days of infection, 10% met the criteria for long COVID at 6 months.
The odds of experiencing 37 symptoms were at least 50% higher among COVID-19 survivors than among uninfected participants (prevalence 2.5% or higher). Hallmark long-COVID symptoms were, by decreasing frequency, post-exertion malaise, fatigue, brain fog, dizziness, gastrointestinal (GI) symptoms, heart palpitations, changes in sexual desire or capacity, loss of or change in smell or taste, thirst, chronic cough, chest pain, and abnormal body movement.
The proportion with a qualifying long-COVID score was 23% among infected participants and 3.7% in uninfected adults. Among participants with long COVID, the most common symptoms were post-exertion fatigue (87%), fatigue (85%), brain fog (64%), dizziness (62%), GI symptoms (59%), and heart palpitations (57%). Higher scores were linked to worse self-reported physical, mental, and social health.
"This study found that long-term symptoms associated with SARS-CoV-2 infection spanned multiple organ systems," the researchers wrote. "The diversity of symptoms may be related to persistent viral reservoirs, autoimmunity, or direct differential organ injury."
The rate of long COVID was lower among fully vaccinated than among unvaccinated participants, and those reinfected with the Omicron variant were more likely to have long COVID than those infected only once.
'A syndrome of syndromes'
"Now that we're able to identify people with long COVID, we can begin doing more in-depth studies to understand the biological mechanisms at play," corresponding author Andrea Foulkes, ScD, of MGH, said in a Mass General Brigham news release.
"One of the big takeaways from this study is the heterogeneity of long COVID: long COVID is not just one syndrome; it's a syndrome of syndromes," she added. "Understanding this idea is a really important step for doing more research and ultimately administering informed interventions."
First author Tanayott Thaweethai, PhD, of Harvard Medical School and MGH, said he hopes the new definition will boost clinical awareness of long COVID. "This is a truly data-driven approach to defining long COVID as a new syndrome," he said in the release.
One of the big takeaways from this study is the heterogeneity of long COVID: long COVID is not just one syndrome; it's a syndrome of syndromes.
The authors emphasized the preliminary nature of their definition. "As a first step to providing a framework for other investigations, iterative refinement that further incorporates other clinical features is needed to support actionable definitions of PASC," they wrote.
"Definition of a classification rule for PASC requires an updated algorithm that incorporates symptoms as well as biological features," they concluded. "Future analyses must consider the relationships among age, sex, race and ethnicity, social determinants of health, vaccination status after index date, comorbidities, and pregnancy status during infection on the risk of PASC and the distribution of PASC subgroups."