Algorithm dependence of patient phenotypes in Long COVID: a patient-led, multi-method clustering of 6031 patients using 162 self-reported symptoms

Tessa D Green, Chris McWilliams, Leonardo de Figueiredo, Letícia Soares, Beth Pollack, Alison K Cohen, Tan Zhi-Xuan, Tess Falor, Hannah E Davis · 2026

Read Algorithm dependence of patient phenotypes in Long COVID: a patient-led, multi-method clustering of 6031 patients using 162 self-reported symptoms by Tessa D Green, Chris McWilliams, Leonardo de Figueiredo, Letícia Soares, Beth Pollack, Alison K Cohen, Tan Zhi-Xuan, Tess Falor, Hannah E Davis (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13284999.txt.

Description

Tessa D. Green and co-authors test how method choice shapes Long COVID patient groups in a 2026 patient-led study. Long COVID symptoms vary widely, and past work has named many symptom clusters now used in trials. The team ran three clustering methods on the same survey data from 6,031 adults with at least ninety days of illness.

Each method gave plausible groups, but agreement between methods was low in the authors analysis. All three found one high-burden group with strong post-exertional malaise, and lower-burden groups that were older with fewer women. Wider mapping showed symptoms blend in a smooth range, with no clear borders. The authors warn single-method groups may be unstable.

Catalog record

CREATORTessa D Green, Chris McWilliams, Leonardo de Figueiredo, Letícia Soares, Beth Pollack, Alison K Cohen, Tan Zhi-Xuan, Tess Falor, Hannah E Davis
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57367
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13284999

Browse the full library · About the Tower of Babel