Abstract Background Diabetes mellitus (DM) is a growing global health burden, yet early identification of at-risk individuals remains challenging. Exercise, a key indicator of cardiovascular and metabolic health, may provide valuable insights into DM risk. This study aimed to evaluate the association between exercise capacity and new-onset DM, as well as its impact on mortality and major adverse cardiovascular events (MACE). Methods We conducted a retrospective cohort analysis of individuals who underwent treadmill testing between 2003 and 2012. Exercise capacity, measured in metabolic equivalents (METs), and blood pressure responses during exercise were recorded. Participants with preexisting DM or major illnesses were excluded. Incident DM was identified during follow-up. Cox proportional hazards models were used to assess the association between exercise metrics and DM risk. Additionally, DM was treated as a time-dependent variable, and Cox regression was employed to evaluate the relationship between METs, DM, mortality, and MACE. Results Over a median follow-up of 8.8 years, 712 (6.65%) of the 10,706 participants developed DM. Compared to those who remained free from DM, individuals with incident DM had lower peak METs (9.36 vs. 10.10, p 0.001), higher body mass index, and poorer metabolic profiles. Lower peak METs independently predicted new-onset DM, with a hazard ratio (HR) of 0.848 per 1-MET increase (95% confidence interval CI: 0.819–0.878), and a nonlinear relationship was observed. Additionally, resting systolic blood pressure (SBP) and pulse pressure (PP) were also significant predictors of incident DM (HR: 1.010, 95% CI: 1.004-1.015; and HR: 1.013, 95% CI: 1.005-1.020, respectively). Stepwise regression analysis indicated that peak METs was a stronger predictor of DM than resting SBP or PP. When DM was treated as a time-dependent variable in multivariate regression analysis, peak METs remained independently associated with a lower risk of all-cause mortality (HR: 0.831, 95% CI: 0.768–0.898) and MACE (HR: 0.847, 95% CI: 0.816–0.879). Resting PP was also an independent predictor of MACE. Conclusions Lower exercise capacity is a strong independent predictor of new-onset diabetes, all-cause mortality, and MACE. The inverse relationship between peak METs and DM risk is nonlinear, emphasizing the protective role of higher METs. While resting SBP and PP were also associated with incident DM, exercise capacity remained the strongest predictor. Even after DM onset, higher METs correlated with lower mortality and MACE risk. These findings support incorporating exercise capacity assessment into routine clinical practice for early diabetes risk stratification and cardiovascular prevention.relationship of incident DM with METs DM as a time-dependent covariate
Wu et al. (2025) studied this question.