Aims <p>To assess the effect of a 1-year supervised exercise training program on long-term mortality in people with type 2 diabetes.</p> Methods <p>This is a post hoc analysis of the Italian Diabetes and Exercise Study randomized clinical trial, which enrolled 606 physically inactive and sedentary individuals with type 2 diabetes from October 2005 to March 2006. Participants were randomized 1:1 to either an exercise group (EXE) receiving a twice-weekly, progressive, supervised aerobic and resistance exercise training plus a physical activity (PA) counseling or a control group (CON) receiving counseling alone for one year. The vital status of participants was verified on 30 June 2024 by interrogating the Italian Health Card database.</p> Results <p>A similar number of EXE and CON participants died over a mean 16.2-year follow-up (88 vs. 95, <i>p</i> = 0.536). Likewise, no between-group differences were detected in death rates (unadjusted, 17.62 [95% confidence interval, 14.30-21.71] vs. 19.59 [16.02–23.96] per 1,000 patient-years, <i>p</i> = 0.483; and age- and sex-adjusted, 16.46 [12.81–21.13] vs. 17.93 [13.92–23.09], <i>p</i> = 0.563). Kaplan-Meier estimates (Log Rank = 0.647, <i>p</i> = 0.421) and mortality risk (unadjusted, hazard ratio, 0.888 [95% confidence interval, 0.664–1.187], <i>p</i> = 0.442; and age- and sex-adjusted, 0.922 [0.689–1.233], <i>p</i> = 0.584). A significant relation with mortality was observed for baseline maximal oxygen uptake (VO_2max), but not lower body muscle strength, regardless of study arm, whereas no association was detected for changes from baseline in leisure-time PA, total PA volume, VO_2max, and lower body muscle strength.</p> Conclusions <p>In people with type 2 diabetes, a one-year supervised exercise training program had no significant effect on long-term mortality.</p> Trial registration <p>isrctn.com; ISRCTN-04252749; URL <a href="https://www.isrctn.com/ISRCTN04252749">https://www.isrctn.com/ISRCTN04252749</a>.</p>