Investigators updated the classification system for long COVID, identifying five distinct symptom patterns and establishing new diagnostic thresholds, according to findings from the National Institutes of Health's RECOVER Initiative.
In the 2024 update, published in JAMA, the investigators analyzed the data from 13,647 participants across 83 U.S. sites and Puerto Rico. The study found that 20% of the participants with prior SARS-CoV-2 infections met the criteria for likely long COVID, whereas 39% of them qualified for a new "possible long COVID" classification.
The updated diagnostic model identified 11 key symptoms: postexertional malaise, fatigue, brain fog, dizziness, palpitations, changes in smell or taste, thirst, chronic cough, chest pain, shortness of breath, and sleep apnea. A score of 11 or higher on these symptoms indicated likely long COVID.
The analysis revealed 5 symptom patterns:
- Subtype 1: Changes in smell or taste
- Subtype 2: Chronic cough
- Subtype 3: Brain fog
- Subtype 4: Palpitations
- Subtype 5: Multiple symptoms including postexertional soreness, dizziness, and gastrointestinal issues.
Fatigue and postexertional malaise appeared prominently across all subtypes except subtype 1. Participants with subtype 5 reported lower quality of life and daily function scores.
The data showed female patients experienced subtypes 4 and 5 more frequently, whereas Hispanic and multiracial participants had higher rates of subtype 5. Unvaccinated individuals and those infected prior to Omicron variant circulation showed higher rates of subtype 5.
The updated index demonstrated correlation with the 2023 version, with 18.8% of the participants meeting both thresholds. The investigators noted that the index was designed for research purposes rather than clinical diagnosis.
Study limitations included the inability to distinguish time-dependent effects on symptoms, potential misclassification of uninfected participants as a result of decreased testing and waning antibodies, and possible confounding from other medical conditions and medications.
Conflict of interest disclosures can be found in the study.