Gellish et al.
About this reference
Gellish and colleagues retrospectively analysed 908 maximal graded exercise tests from 132 participants tested repeatedly at a university fitness centre over 25 years. The longitudinal mixed model produced 207 - 0.7 x age. FitnessNorms cites this as a design-based comparison with Tanaka; it does not drive the displayed percentile grid.
Known limitations
- Only 132 unique participants; 908 is the number of repeated tests, not the participant count
- Fitness-centre members may not represent the general population
- Single US centre limits generalisability