Introduction It is unknown if short-term worsening in COPD is a predictor of long-term poor outcome. We assessed whether worsening measured using a short-term composite measure (clinically important deterioration [CID]) would predict adverse outcomes in two large 3-year studies. Methods A CID was defined as any of: a decrease of ≥100mL in FEV1 or increase of ≥4 units in St George9s Respiratory Questionnaire (SGRQ) score from baseline or a moderate/severe exacerbation at ≤6 months (index date [ID]) in TORCH and ≤12-months ID in ECLIPSE. Association between presence (+) and absence (-) of CID status at ID and long-term deterioration in FEV1, health-status, future risk of exacerbations and all-cause mortality was assessed post hoc from ID until end of follow-up. Only subjects that did not withdraw before the ID were included in the analysis. Results In total, 2870 of 5292 (54%) and 1442 of 1973 (73%) patients with data post ID were CID+ in TORCH and ECLIPSE, respectively. In both studies, CID+ patients had a clinically significant deficit in FEV1 and health-status and higher exacerbation risk (p<0.001 vs CID- group). All-cause mortality was also higher (p<0.05 vs CID- patients). Conclusion A CID that occurs within 6–12 months of follow-up is associated with sustained loss of lung function and health-status and increased exacerbation and all-cause mortality risk. Funded by GSK (NCT00268216, SCO30003; NCT00292552, SCO104960).