Maximal heart rate declines linearly with age independent of cardiorespiratory fitness levels.

There have been many conflicting observations between the linear or curvilinear decline in maximal heart rate (HR) with age. The aim of this study was to determine if linear or curvilinear equations would better describe the decline in HR with age in individuals of differing cardiorespiratory fitness (CRF) levels. Treadmill cardiopulmonary exercise test (CPX) results from participants (1510 men and 1134 women; 18-76 years) free of overt cardiovascular disease were retrospectively examined using cross-sectional and longitudinal study designs. Participants completing ≥2 CPX with ≥1 year between test dates were included in the longitudinal analysis (325 men and 150 women). Linear and quadratic regressions were applied to age and HR for the whole cohort and respective CRF groups (high, moderate, and low, relative to age and gender normative values). To test for differences among linear, quadratic, and polynomial equations, the change in R (cross-sectional analysis) and Bayesian information criterion (BIC) (longitudinal analysis) from the linear to the more complex models were calculated. The quadratic or polynomial regression in the cross-sectional analysis, marginally improved the variance in HR explained by age compared to the linear regression for the whole cohort (0.2%), moderate fit group (0.3%), and low fit group (0.8%). With no improvements in the high fit group. BIC did not improve for any CRF category in the longitudinal analysis. In conclusion, the minimal differences among linear, quadratic, and polynomial equations in the respective CRF groups, emphasizes the use of linear prediction equations to estimate HR.