Patients taking glucagon-like peptide-1 receptor agonists showed a lower risk of tuberculosis than those prescribed other standard blood-sugar medications, according to a study published in Nature Communications.
Researchers analyzed electronic health records from 143 healthcare organizations across multiple countries using the TriNetX network. The cohort study evaluated clinical outcomes between 2017 and 2025, applying propensity score matching and time-to-event analyses across follow-up periods reaching up to five years.
Patients prescribed GLP-1 receptor agonists experienced lower incidence rates of tuberculosis than those taking alternative glucose-lowering drugs. When compared against sulfonylureas, the GLP-1 group had an incidence rate of 0.68 cases per 1,000 person-years compared to 1.67 for sulfonylureas, yielding a hazard ratio of 0.53 with a 95 percent confidence interval between 0.47 and 0.59. Relative to metformin, GLP-1 patients recorded 0.65 cases per 1,000 person-years versus 1.31, reflecting a hazard ratio of 0.60 and a confidence interval between 0.51 and 0.70.
Comparisons with dipeptidyl peptidase-4 inhibitors revealed incidence rates of 0.73 per 1,000 person-years for GLP-1 users versus 1.71 for DPP-4 inhibitors, producing a hazard ratio of 0.49 with a confidence interval between 0.43 and 0.56. When measured against sodium-glucose cotransporter-2 inhibitors, GLP-1 users recorded 0.89 cases per 1,000 person-years against 1.02, reflecting a hazard ratio of 0.82 with a confidence interval of 0.72 to 0.92.
Researchers K.M. Liao, J.Y. Wu, and C.C. Lai led the investigation across institutions including Chi Mei Hospital, Min-Hwei Junior College of Health Care Management, Chi Mei Medical Center, National Cheng Kung University, and National Sun Yat-sen University in Taiwan. The authors declared no competing interests and no relevant funding for the study, which Nature Communications received on August 19, 2025, accepted on August 11, 2026, and published on August 22, 2026.
