We tested the validity of an new methodological approach to the calculation of oxygen uptake efficiency slope (OUES) [i.e. the use of exercise end-points based on fractions of heart rate reserve (HR_res)], as an alternative to the traditional time-based calculation. Twenty-nine healthy sedentaries >60 years of age (18 males, 11 females) performed an incremental cycling exercise to exhaustion. Respiratory variables and HR were measured breath by breath. Resting and peak variables were calculated and ventilatory threshold (VT) was identified by standard technique. OUES was calculated on 75, 90 and 100% of the incremental exercise data (OUES75, OUES90, OUES100) and on data corresponding to 60 and 80% of the HR_res (OUES60%HR_res, OUES80%HR_res). No significant difference (repeated measures ANOVA) was detected between time-based (OUES100, OUES90, OUES75) as well as HR_res-based measures of OUES (OUES80%HR_res, OUES60%HR_res). The Bland–Altman analysis revealed a bias not significantly different from 0 (22.0 and 53.3 for OUES80%HR_res-OUES100 and OUES60%HR_res-OUES100, respectively), a precision of 171.2 and 289.0 and 95% limits of agreement from −313 to +358 and from −513 to +620 for OUES80%HR_res-OUES100 and for OUES60%HR_res-OUES100, respectively. High correlations were detected between ( [(V)\dot]\textO_\text2peak ) {\left( {\dot{V}{\text{O}}_{{{\text{2peak}}}} } \right)} and OUES60%_res and OUES80%HR_res (r ² = 0.70 and 0.81, respectively) and between VT and OUES60%_res and OUES80%HR_res (r ² = 0.58 and 0.66, respectively). The main finding of this study is that OUES can be reliably calculated based on HR_res end-points during incremental cycling exercise, in healthy elderly subjects. Furthermore, our study confirms the validity of OUES as an indicator of aerobic exercise capability in this population.