Putting the PASC Score to the Test: Clinical vs. Statistical Accuracy in Long COVID Diagnosis.
Azola, Alba, Dastgheyb, Raha M, Easter, Rebecca et al. · Journal of general internal medicine · 2025 · DOI
Quick Summary
Researchers tested a scoring tool called the PASC score to see if it could correctly identify long COVID patients. They compared 130 long COVID patients with 60 people who recovered completely from COVID-19. The PASC score was very accurate at confirming who had long COVID, but missed about 20% of actual patients. A simpler approach using just three symptoms (loss of smell/taste, post-exertional malaise, or sexual dysfunction) worked even better at catching long COVID cases.
Why It Matters
Accurate diagnostic tools are essential for long COVID and ME/CFS identification, as both conditions currently lack biomarkers. This research provides evidence for a validated screening instrument and suggests that a simpler three-symptom approach may be more clinically practical, potentially improving access to diagnosis and care for affected patients.
Observed Findings
The PASC score showed 100% specificity and positive predictive value but only 80% sensitivity in identifying long COVID using NASEM criteria.
A three-symptom combination (loss of smell/taste, post-exertional malaise, or sexual dysfunction) demonstrated 94% sensitivity and 92% specificity.
The LC cohort was predominantly female (72%), White (79%), mean age 47.2 years, and well-educated (77% with >16 years education).
LC diagnosis and PASC scores were significantly associated (χ²=102.99, P<0.001).
The three-symptom approach achieved an F1 score of 0.949, indicating excellent overall diagnostic accuracy.
Inferred Conclusions
The PASC score is a valid tool for long COVID diagnosis but the current cutoff (>12) misses approximately 20% of affected individuals.
A simplified three-symptom screening approach may be more clinically useful than the full PASC score for identifying long COVID cases.
Ongoing refinement of both the long COVID definition and diagnostic tools is necessary to improve case identification and clinical utility.
Remaining Questions
How does the PASC score perform in populations with different demographic characteristics, particularly non-White and lower-education groups not well-represented in this cohort?
What This Study Does Not Prove
This study does not establish causation or understand why these specific symptoms are associated with long COVID. Being cross-sectional, it captures a single time point and cannot determine how the PASC score performs in early disease or whether it predicts long-term outcomes. The study's demographic homogeneity (79% White, 77% college-educated) may limit generalizability to more diverse populations.
About the PEM badge: “PEM required” means post-exertional malaise was an explicit required diagnostic criterion for participant inclusion in this study — not that PEM was studied, observed, or discussed. Studies using criteria that do not require PEM (e.g. Fukuda, Oxford) are tagged “PEM not required”. How the atlas works →
Contribute
Private, reviewed by a human. Not a public comment thread.