E2 ModerateModerate confidencePEM ?Cross-SectionalPeer-reviewedMachine draft
Psychometric evaluation of the DePaul Symptom Questionnaire-Short Form (DSQ-SF) among adults with Long COVID, ME/CFS, and healthy controls: A machine learning approach.
McGarrigle, William J, Furst, Jacob, Jason, Leonard A · Journal of health psychology · 2024 · DOI
Quick Summary
Researchers tested whether a symptom questionnaire called the DSQ-SF can accurately identify and distinguish between Long COVID, ME/CFS, and healthy people. Using computer algorithms to analyze the results, they found the questionnaire works well at telling these groups apart and identified which specific symptoms are most helpful for telling Long COVID and ME/CFS apart from each other and from normal health.
Why It Matters
Accurate and efficient screening tools are essential for improving diagnosis and treatment of both ME/CFS and Long COVID. This research validates the DSQ-SF as a reliable instrument for distinguishing these overlapping conditions, which could accelerate clinical recognition and enable more targeted therapeutic approaches for affected patients.
Observed Findings
- The DSQ-SF demonstrated high sensitivity and specificity in classifying Long COVID, ME/CFS, and healthy control groups using machine learning algorithms.
- Specific DSQ-SF symptom items were identified that best distinguish Long COVID from ME/CFS.
- Other DSQ-SF items were found to effectively differentiate both illness groups from healthy controls.
Inferred Conclusions
- The DSQ-SF is a validated and efficient tool for screening and classifying adults with Long COVID and ME/CFS.
- Certain symptom patterns on the DSQ-SF can help clinicians differentiate between Long COVID and ME/CFS despite their clinical overlap.
- Machine learning approaches can enhance the psychometric evaluation and clinical utility of symptom assessment questionnaires.
Remaining Questions
- How does the DSQ-SF perform across different demographic groups, disease severity levels, and stages of illness?
- Can the DSQ-SF predict treatment response or prognosis in Long COVID and ME/CFS patients?
- How stable are DSQ-SF scores over time, and can the questionnaire detect meaningful changes in symptom burden with treatment?
- What is the biological or mechanistic basis for the symptom differences the DSQ-SF identifies between Long COVID and ME/CFS?
What This Study Does Not Prove
This study does not prove that the DSQ-SF is superior to other diagnostic instruments or that it should replace medical evaluation and clinical assessment. The study also does not establish the DSQ-SF's ability to predict disease progression, treatment response, or long-term outcomes, nor does it determine whether identified symptom differences reflect underlying biological distinctions between Long COVID and ME/CFS.
Tags
Symptom:Post-Exertional MalaiseCognitive DysfunctionFatigue
Phenotype:Long COVID Overlap
Method Flag:PEM Not DefinedStrong Phenotyping