Derivation and validation of a pre-bronchodilator FEV1/FVC threshold for point-of-care diagnosis triage of COPD in primary care: a multi-cohort population-based study.

Background: Post-bronchodilator (BD) spirometry is underused in primary care settings, limiting COPD diagnosis. This study aimed to derive and validate pre-BD thresholds that safely rule in COPD and enable immediate diagnosis. Methods: We derived pre-BD FEV1/FVC thresholds using a nationally representative derivation dataset (CPHS, n = 6360 with pre-BD ratio <0.7), with cross-sectional external validation in the China Kadoorie Biobank (CKB, n = 1890) and longitudinal validation in a national screening cohort (NSC, n = 7620 with 1-year follow-up). The optimal threshold was selected based on positive predictive value (PPV) for post-BD FEV1/FVC <0.7. One-year longitudinal follow-up assessed diagnostic stability and prognostic value. Findings: A pre-BD threshold of 0.56 was identified, yielding a PPV of 0.953 (95% CI, 0.939-0.965) in derivation set and 0.919 (95% CI, 0.875-0.940) in the CKB validation set. In subgroups with higher prior probability (male, age >60 years, smoking, and with respiratory symptoms), the PPV was further increased. At one-year follow-up in the NSC, 86.9% of participants with baseline pre-BD <0.56 still met the post-BD COPD criteria. Patients below 0.56 had significantly higher risks of acute exacerbation (OR, 3.34; 95% CI, 2.55-4.37) and respiratory hospitalization (OR, 3.42; 95% CI, 2.65-4.42). An alternative threshold of 0.62 gave a PPV of 0.891 (95% CI, 0.866-0.913) in CKB and allowed 33.8%-49.0% of participants across cohorts to bypass post-BD testing. Interpretation: A pre-BD FEV1/FVC threshold of 0.56 effectively "rules-in" COPD in primary care and identifies patients with poor prognosis. The alternative threshold of 0.62 offers a practical option to further reduce post-BD testing while maintaining an acceptable PPV. Both thresholds facilitate pragmatic triage and early diagnosis without routine post-BD testing. Funding: This work was funded by National High Level Hospital Clinical Research Funding (2025-NHLHCRF-JBGS-B-WZ-18), Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0506003, 2023ZD0506306, 2023ZD0510101, 2023ZD0510100), Social Development Project of Yunnan Province (202403AC100006) and Beijing Nova Program (20250484849).