Prediction Equations for Spirometry for Children from Northern India

Area of Science:

• Pulmonary Medicine
• Pediatric Respiratory Health
• Biostatistics

Background:

• Spirometry is crucial for assessing lung function in children.
• Existing spirometry prediction equations may not accurately represent diverse pediatric populations, including those from northern India.
• Standardization of spirometry testing is essential for reliable data.

Purpose of the Study:

• To develop and validate new spirometry prediction equations for healthy children aged 6-17 years residing in northern India.
• To adhere to current international guidelines for spirometry standardization.
• To compare the performance of linear and nonlinear regression models for predicting lung function parameters.

Main Methods:

• Re-analysis of cross-sectional spirometry data from 670 normal children (365 boys) in northern India.
• Spirometry performed according to current international guidelines with quality assurance.
• Linear and nonlinear multiple regression analyses used to develop prediction equations for FVC, FEV1, PEFR, FEF50, FEF75, and FEF25-75.
• Model selection based on the highest coefficient of multiple determination (R2) and statistical validity.

Main Results:

• Prediction equations were derived for forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV1) for boys and girls.
• Nonlinear regression models generally yielded higher R2 values than linear models, indicating improved prediction accuracy, except for FEF50 in girls.
• Height and age were significant predictors for all spirometry parameters, while weight did not contribute significantly.

Conclusions:

• Novel spirometry prediction equations have been successfully developed for the pediatric population of northern India.
• Nonlinear regression models provide a more accurate representation of lung function compared to linear models in this cohort.
• These equations will aid in the standardized assessment of respiratory health in Indian children.