Assessing the Validity of Two Non-Exercise Regression Equations for Predicting Maximal Oxygen Consumption

Purpose: This study aimed to develop two regression equations to predict maximal oxygen consumption (VO_2max) using non-exercise data from a substantial cohort of healthy Iranian adult males. Additionally, this study sought to examine the predictive accuracy of these equations across four different levels of physical activity. Methods: A total of 126 participants (age: 34.9 ± 11.3 years, body mass index [BMI]: 24.9 ± 2.7 kg/m², and body fat percentage [BF%]: 18.3 ± 4.9) completed a maximal graded exercise test to measure VO_2max, with a mean of 45.0 ± 3.4 ml.kg^-1.min^-1. Participants also provided information on age, current physical activity rating (PA-R), and either BMI or BF% to estimate VO_2max using Jackson and colleagues' regression equations. The PA-R was assessed via a standardized questionnaire and categorized into four levels: sedentary, low, moderate, and high. Results: The key findings from this study indicate that both original models significantly underestimated actual VO_2max in a large cohort of Iranian adults (both, p < .001 and mean differences exceeding 2.19 ml.kg^-1.min^-1). Nevertheless, these models provided accurate predictions for VO_2max among individuals with moderate levels of physical activity (both, p > .08 and mean differences between 0.51 and 1.03 ml.kg^-1.min^-1). Furthermore, the models demonstrated moderate validity, as evidenced by an intraclass correlation coefficient (ICC) of 0.841 and a coefficient of variation averaging 10.9%, with a range from 8.5% to 13.6%. Conclusions: While Jackson's two non-exercise models showed limited accuracy in predicting VO_2max among Iranian healthy male adults, they exhibited reasonable precision, particularly among moderately active men.