Stair-Climbing Versus Machine-Based Resistance Exercise to Improve Muscle Power Among Older Adults: A Noninferiority Trial

Abstract Van Roie, E, van Uffelen, J, and Delecluse, C. Stair-climbing versus machine-based resistance exercise to improve muscle power among older adults: a non-inferiority trial. J Strength Cond Res 39(3): e496–e505, 2025—Machine-based resistance training (RT) can reduce the age-related loss in muscle power ( P max ). However, weight-bearing exercises have greater potential for implementation. This study investigated whether stair-climbing exercise (STAIR) was non-inferior to improve P max compared with machine-based RT among older adults. Functional capacity tests were secondary outcomes. Older adults (30♂ and 16♀; 70.9 ± 4.3 years) were randomly assigned to RT or STAIR ( n = 23 per group). Supervised laboratory-based training sessions were performed 2 times per week for 12 weeks. In week 1–4, exercises were performed at controlled speed (hypertrophy-oriented; 4 × 12–15 repetitions; 55% of 1 repetition maximum (1RM) in RT; step-up exercise with height of 30–40 cm in STAIR), in week 5–12 as fast as possible (power-oriented, 4 × 12 repetitions; 40% of 1RM for RT and 4 × 2 flights of 6 steps for STAIR). Leg-extensor P max and functional capacity were measured pre-intervention and post-intervention, and P max also after 4 weeks of training. Gains in P max were not statistically different in RT (19.5 ± 12.2%) and STAIR (13.7 ± 16.5%) ( d = 0.39, p int = 0.086), whereas non-inferiority analyses were inconclusive (between-group difference in gains: 5.8% (95% CI −3.1 to 14.6); non-inferiority limit 3.9%). STAIR increased more in stair ascent performance ( d = 0.45–0.61, p int < 0.05) than RT. STAIR and RT improved similarly on 10-m fast walk, 5-repetition sit-to-stand, and countermovement jump ( p int > 0.05). To conclude, STAIR and RT both induced significant changes in P max and functional capacity in older adults. STAIR was beneficial over RT for gains in stair-climbing performance. Future research should investigate the effects of STAIR in a home-based unsupervised setting.