Background Cardiorespiratory fitness (CRF) has important implications for post-operative recovery. The timedup-and-go (TUG) test is a cheap and simple method to assess a patient's functional performance; although how well TUG correlates with results of a cardiopulmonary exercise test (CPET), the gold standard measure of CRF is unknown. Therefore, the aim of this study was to assess the correlation between CPET-derived parameters of CRF and TUG times in a group of older adults. Methods Ninety-eight independent community dwelling older adults [mean age: 72 years (range: 61-86), mean body mass index: 26.3 ± 3.1 kg/m 2 , 54 male] were recruited to this study; completing 180 CPET and TUG testing sessions over a 28 month period. The correlation between CPET-derived CRF parameters and TUG time was assessed, and receiver operating characteristic curve analysis was performed to determine clinically useful cut-off points in TUG time. Results Median TUG time was 7.1 s [interquartile range (IQR): 4-8.5], median VO 2 peak was 24.4 mL/kg/min (IQR: 20.2-29.2), and the median anaerobic threshold (AT) was 13.4 mL/kg/min (IQR: 8.6-16.5). There was a statistically significant negative correlation between TUG time and AT (r = À0.317, P = <0.0001) and TUG time and VO 2 peak (r = À0.4247, P < 0.0001). Receiver operating characteristic curve analysis determined a TUG time of ≥6.5 s to have an 82% sensitivity and 60% specificity to detect an AT <11.0 mL/kg/min, the point at below which perioperative mortality is known to increase. Conclusions Despite strong evidence for the utility of pre-operative CPET in stratifying surgical risk, CPET is not universally available. Our finding of a correlation between TUG time and CPET-derived parameters of CRF (AT/VO 2 peak) suggests that TUG may be a useful surrogate in the pre-operative setting.