Long COVID affects twice as many Americans as official counts show, new AI study finds

A study published in *JAMA Network Open* reveals that long COVID affects nearly one in six Americans who had COVID-19, far exceeding the 7% detected by standard billing codes. Researchers from Mass General Brigham used an AI algorithm, P2RC, to analyze electronic health records of 457,950 patients across 58 hospitals in four U.S. regions. By examining full medical histories, they identified chronic symptoms unexplained by preexisting conditions, with regional rates ranging from 13.6% in Western Pennsylvania to 22.7% in Southern California. The findings suggest over 10 million long COVID cases go undetected due to reliance on outdated diagnostic codes. Lead author Jiazi Tian noted that doctors often see long COVID symptoms—such as heart, metabolic, or neurological issues—without linking them to prior COVID-19 infections. Corresponding author Hossein Estiri warned the true prevalence may be even higher, as the study’s data only extends to mid-2024. The study also found that 14.5% of infected patients developed new chronic conditions requiring ongoing care, including diabetes, heart disease, and cognitive impairment. Researchers emphasized long COVID should be treated as a chronic disease burden, not a temporary syndrome. Cumulative prevalence continued rising through mid-2024, with projections indicating sustained growth if current trends persist. The results align with prior research, including a 2023 study by Dr. Ziyad Al-Aly at the Department of Veterans Affairs, which linked COVID-19 to long-term organ damage even after mild infections. The NIH’s RECOVER initiative and international studies similarly estimate long COVID prevalence between 10% and 25% among infected adults, with many remaining severely ill years later. The Brigham study’s 16.3% figure falls within this range and highlights severe metabolic and endocrine consequences. The study underscores a critical gap in public health surveillance, as current diagnostic methods fail to capture the full scope of long COVID. Researchers warn that undercounting impedes clinical management and may contribute to rising post-pandemic rates of chronic diseases. The findings call for improved diagnostic tools and coordinated healthcare responses to address the growing burden of long COVID.