BACKGROUND: Evidence has shown benefits for mental health through aerobic training oriented in percentage of VO2max, indicating the importance of this variable for clinical practice. OBJECTIVE: To validate a method for estimating VO2max using a submaximal protocol in elderly patients with clinically diagnosis as major depressive disorder (MDD) and Parkinson's disease (PD). METHODS: The sample comprised 18 patients (64.22 ± 9.92 years) with MDD (n = 7) and with PD (n = 11). Three evaluations were performed: I) disease staging, II) direct measurement of VO2max and III) submaximal exercise test. Linear regression was performed to verify the accuracy of estimation in VO2max established in ergospirometry and the predicted VO2max from the submaximal test measurement. We also analyzed the correlation between the Bland-Altman procedures. RESULTS: The regression analysis showed that VO2max values estimated by submaximal protocol associated with the VO2max measured, both in absolute values (R² = 0.65; SEE = 0.26; p < 0.001) and the relative (R² = 0.56; SEE = 3.70; p < 0.001). The Bland-Altman plots for analysis of agreement of showed a good correlation between the two measures. DISCUSSION: The VO2max predicted by submaximal protocol demonstrated satisfactory criterion validity and simple execution compared to ergospirometry.