Development and validation of cardiorespiratory fitness prediction equations from 6-minute walk test: the Coronary Artery Risk Development in Adults (CARDIA) study

TL;DR

Researchers developed new equations to estimate cardiorespiratory fitness using a 6-minute walk test, which is easier to perform than traditional methods.

Contribution

The study introduces sex-specific prediction equations for VO2max and GXTd from 6MWT data, improving accuracy over existing models.

Findings

The CARDIA equations explained 53% and 57% of VO2max variance in men and women, respectively.

Compared to existing equations, CARDIA models showed stronger correlations and less bias.

Abstract

Assessing maximal oxygen consumption (VO2max), the gold standard for assessing cardiorespiratory fitness, is often impractical in large-scale studies. We derived sex-specific VO2 max and graded exercise test duration (GXTd) prediction equations from 6-minute walk test (6MWT) performance.

Data were from 564 Coronary Artery Risk Development in Young Adults (CARDIA) participants (mean age: 61.5 years; 58% women; 40% Black) who completed the 6MWT and symptom-limited modified Balke treadmill graded exercise test at the Year 35 (2021-2022) follow-up exam. Sex-stratified samples were randomly split (2/3 for training, 1/3 for testing) to derive and evaluate prediction equations. Stepwise linear regression identified predictors of VO2max and GXTd from 6MWT distance. Sex-specific VO2max CARDIA equations were compared with Burr and FRIEND equations. Models’ accuracies were evaluated by comparing the predicted values to measured values using Student’s t-test and Pearson correlation coefficients (r). Bland−Altman plots used to evaluate agreement between measured and predicted values.

Sex-specific VO2max CARDIA equations explained 53% and 57% of the variance in men and women, respectively, with strong correlations between measured and predicted values (r = 0.73 and 0.78). The Burr and FRIEND equations showed larger biases and weaker correlations compared with the CARDIA equations. The GXTd equations explained 59% and 62% of the variance in men and women, with strong correlations (r = 0.71 and 0.72) and no significant mean differences between observed and predicted.

The CARDIA prediction equations for VO2max and GXTd from 6MWT enhance accuracy and accessibility, providing a practical tool for large-scale studies and clinical assessments, particularly in aging populations.

Figures