ME/CFS Prevalence in the Long COVID Era: A Corrected U.S. Burden Estimate Using Infection-Associated Chronic Condition Modeling

Abstract

Background: ME/CFS is a multi-system neuroimmune condition whose true prevalence has been systematically underestimated for decades. Pre-pandemic U.S. estimates placed burden at 1 to 2.5 million individuals, shaped by narrow diagnostic criteria, fragmented care systems, limited clinical recognition, and structural undercounting. The National Academies of Sciences estimated that 84% to 91% of individuals with ME/CFS remain undiagnosed, establishing long-standing acknowledgment of a large invisible population.

Methods: This is a modeling study analyzing existing population-level data. The US-CCUC™ (U.S. Corrected Chronic Undercounting Correction) framework reconstructs ME/CFS prevalence by integrating multiple documented sources of structural undercount: hidden pre-pandemic cases, post-infectious expansion, demographic under capture, diagnostic misclassification, and remission-state exclusion. The model applies ME/CFS-concordant trajectory rates (40–51%) to an updated Long COVID population baseline and adds a legacy diagnosed ME/CFS denominator, adjusting for overlap and diagnostic uncertainty. Core data inputs include: CDC NHIS 2021–2022 prevalence data (Vahratian et al., 2023); NIH RECOVER-Adult longitudinal cohort findings (Vernon et al., 2025); 2015 Institute of Medicine diagnostic criteria; and published epidemiological literature on underdiagnosis and case definition variability (Jason et al., 2005; Nacul et al., 2011).

Results: The CDC NHIS found 1.3% of U.S. adults reported ME/CFS in 2021–2022, already exceeding prior legacy assumptions. The NIH RECOVER-Adult study found 4.5% of SARS-CoV-2 infected participants met 2015 IOM ME/CFS diagnostic criteria at least six months post-infection, versus 0.6% of uninfected participants (hazard ratio 4.93). Applying ME/CFS-concordant trajectory rates of 40–51% to a Long COVID population baseline of approximately 65 million U.S. adults (derived from RECOVER and CDC longitudinal surveillance data), plus a legacy denominator of approximately 1.5 million diagnosed cases, yields a modeled classification burden of approximately 27.5 to 34.65 million U.S. adults. The conservative public-facing prevalence range, adjusted for overlap, diagnostic uncertainty, and structural invisibility, is estimated at approximately 18 to 26 million U.S. adults (planning midpoint: ~22 million).

Conclusions: Under this corrected modeling framework, ME/CFS represents one of the largest and least visible neuroimmune illnesses in the United States, embedded within a broader infection-associated chronic condition (IACC) landscape estimated to affect 75 to 90 million Americans. Existing prevalence figures reflect diagnostic visibility, not true population burden. These corrected estimates carry direct implications for NIH portfolio planning, clinical education, therapeutic development, disability policy, and economic burden analysis. Independent validation against primary epidemiological cohort data is identified as a priority next step.

Keywords: ME/CFS, myalgic encephalomyelitis, chronic fatigue syndrome, Long COVID, prevalence correction, post-viral illness, neuroimmune, post-exertional malaise, RECOVER, IACC, remission dynamics, modeling study, infection associated chronic conditions