[Application of a multiple linear regression model of FEV1 in pulmonary function test]

Area of Science:

• Pulmonary Medicine
• Biostatistics
• Respiratory Physiology

Background:

• Pulmonary function tests are crucial for diagnosing respiratory diseases.
• Estimating forced expiratory volume in 1 second (FEV1) is vital for assessing lung function.
• Special populations often face challenges with standard pulmonary function tests.

Purpose of the Study:

• To develop a multiple linear regression model for estimating FEV1.
• To enable FEV1 estimation in individuals unable to perform standard pulmonary function tests.
• To validate the model's accuracy in diverse patient groups.

Main Methods:

• Data from 813 individuals undergoing pulmonary function tests were used.
• A multiple linear regression model was constructed using stepwise analysis.
• The model was validated on an independent cohort of 94 individuals.
• Forced oscillation technique (FOT) was used to measure respiratory resistance (Rrs).

Main Results:

• FEV1 showed significant correlations with height, body mass, age (age-dependent), and Rrs.
• The developed model accurately estimated FEV1 in both the modeling and validation groups.
• Calculated FEV1 values closely matched measured FEV1, with high correlation coefficients (r=0.891 and r=0.795 in validation sets).
• Model performance varied slightly based on age groups, with height, gender, age, and Rrs being key predictors for adults, and height for younger individuals.

Conclusions:

• A multiple linear regression model for calculating FEV1 has been successfully constructed.
• The model demonstrates suitability for clinical application, particularly for special populations.
• This approach offers a viable alternative for FEV1 assessment when standard tests are not feasible.