RESUMO
Recent studies have pointed out divergences in the predictive quality of reference equations for respiratory muscle strength, alerting the need to introduce into these equation-specific variables for certain target populations. OBJECTIVES: This study proposes predictive equations of respiratory muscle strength by the variables such as body weight and body mass index (BMI = weight/height2 ). MATERIALS AND METHODS: This is a cross-sectional study, and the evaluation was made by the test of maximum static respiratory pressures (MIP and MEP) according to standard protocol of the Brazilian Society of Pneumology and Tisiology. For data analysis, the following three models of linear regression were adjusted: age, age/weight, age/BMI. The software used in the analysis was the R version 3.2.1. RESULTS: Of the 353 subjects evaluated (229 women and 124 men), 109 subjects were normal weight, 101 subjects were overweight and 143 subjects were obese. The BMI average of the individuals was 31.42 ± 10.26 kg/m2 and age 46.26 ± 16.47 years. The two statistical models that considered the variables weight and BMI had the R2 value of 29.86% for MEP and 21.77% for MIP when the weight was the predictive variable and 21.33% for MIP and 28.38% for MEP when the variable was BMI. CONCLUSION: It was found with the adjusted models that there was a considerable gain in the predictive quality of the models for MEP and MIP adding weight or BMI, without significant difference between both.