2134/20108 Gary O'Donovan Gary O'Donovan Kishan Bakrania Kishan Bakrania Nazim Ghouri Nazim Ghouri Thomas E. Yates Thomas E. Yates L.J. Gray L.J. Gray Mark Hamer Mark Hamer Emmanuel Stamatakis Emmanuel Stamatakis Kamlesh Khunti Kamlesh Khunti Melanie J. Davies Melanie J. Davies Naveed Sattar Naveed Sattar Jason M.R. Gill Jason M.R. Gill Non-exercise equations to estimate fitness in white European and South Asian men Loughborough University 2016 Physical fitness Exercise test Linear models Validation studies Medical and Health Sciences not elsewhere classified 2016-01-21 13:31:47 Journal contribution https://repository.lboro.ac.uk/articles/journal_contribution/Non-exercise_equations_to_estimate_fitness_in_white_European_and_South_Asian_men/9628952 © 2015 American College of Sports Medicine PURPOSE: Cardiorespiratory fitness is a strong, independent predictor of health, whether it is measured in an exercise test or estimated in an equation. The purpose of this study was to develop and validate equations to estimate fitness in middle-aged white European and South Asian men. METHODS: Multiple linear regression models (n=168, including 83 white European and 85 South Asian men) were created using variables that are thought to be important in predicting fitness (VO2 max, mL⋅kg⋅min): age (years); BMI (kg·m); resting heart rate (beats⋅min); smoking status (0=never smoked, 1=ex or current smoker); physical activity expressed as quintiles (0=quintile 1, 1=quintile 2, 2=quintile 3, 3=quintile 4, 4=quintile 5), categories of moderate- to vigorous-intensity physical activity (0=<75 min⋅wk, 1=75-150 min⋅wk, 2=>150-225 min⋅wk, 3=>225-300 min⋅wk, 4=>300 min⋅wk), or minutes of moderate- to vigorous-intensity physical activity (min⋅wk); and, ethnicity (0=South Asian, 1=white). The leave-one-out-cross-validation procedure was used to assess the generalizability and the bootstrap and jackknife resampling techniques were used to estimate the variance and bias of the models. RESULTS: Around 70% of the variance in fitness was explained in models with an ethnicity variable, such as: VO2 max = 77.409 - (age*0.374) – (BMI*0.906) – (ex or current smoker*1.976) + (physical activity quintile coefficient) – (resting heart rate*0.066) + (white ethnicity*8.032), where physical activity quintile 1 is 1, 2 is 1.127, 3 is 1.869, 4 is 3.793, and 5 is 3.029. Only around 50% of the variance was explained in models without an ethnicity variable. All models with an ethnicity variable were generalizable and had low variance and bias. CONCLUSION: These data demonstrate the importance of incorporating ethnicity in non-exercise equations to estimate cardiorespiratory fitness in multi-ethnic populations.