Estimating Prevalence, Demographics, and Costs of ME/CFS Using Large Scale Medical Claims Data and Machine Learning.
Valdez, Ashley R, Hancock, Elizabeth E, Adebayo, Seyi et al. · Frontiers in pediatrics · 2018 · DOI
Quick Summary
Researchers used insurance billing records and computer analysis to estimate how many Americans have ME/CFS. They found that between 1.7 and 3.4 million people in the U.S. may have been diagnosed with ME/CFS, making it more common than many people realize. The study also showed that ME/CFS costs patients far more in medical expenses than other serious diseases like lupus or multiple sclerosis.
Why It Matters
This study provides quantitative evidence that ME/CFS is not rare but a relatively common condition affecting millions of Americans, challenging previous underestimates of prevalence. The finding that ME/CFS generates substantially higher healthcare costs than similarly serious neurological conditions highlights the significant disease burden and supports the need for greater clinical recognition and research investment.
Observed Findings
Diagnosed prevalence of ME/CFS ranged from 519-1,038 per 100,000 people in commercially insured populations.
Females were diagnosed with ME at 1.24 times the rate of males, and with CFS at 1.18 times the rate of males, though 35-40% of diagnosed patients were men.
Average annual healthcare costs for ME/CFS patients were approximately 50% higher than for lupus or multiple sclerosis, and 3-4 times higher than for the general insured population.
A machine learning model predicted ME/CFS prevalence of 857 per 100,000, translating to approximately 2.8 million U.S. patients.
Inferred Conclusions
ME/CFS is not a rare disease but a relatively common condition affecting roughly 2.8 million Americans.
ME/CFS causes substantially greater healthcare expenditures than other serious neurological conditions, indicating significant disease burden.
ME and CFS may represent distinct or partially overlapping conditions, with CFS showing greater heterogeneity than ME.
Current insurance claim diagnostic codes are insufficient to reliably identify ME/CFS cases based solely on symptom information.
Remaining Questions
How many ME/CFS patients remain undiagnosed in the general population, and what barriers prevent diagnosis?
What This Study Does Not Prove
This study does not establish what causes ME/CFS or prove that the conditions diagnosed as ME and CFS are biologically identical—in fact, it suggests CFS may represent a more heterogeneous group. It also cannot determine how many undiagnosed ME/CFS patients exist in the general population, as it only analyzes people who received a diagnosis in insurance records.
Tags
Method Flag:PEM Not DefinedWeak Case DefinitionExploratory OnlyMixed Cohort
What biological or clinical differences distinguish the more heterogeneous CFS diagnosis category from ME?
Could improved clinical coding and diagnostic criteria in insurance systems better capture the true prevalence of ME/CFS?
Why do ME/CFS patients incur such substantially higher healthcare costs compared to other serious chronic conditions, and what specific services drive these expenses?