Predictors of Respiratory Oscillometry Measurements in a Healthy Population

Area of Science:

• Pulmonary Physiology
• Respiratory Medicine
• Biostatistics

Background:

• Impulse oscillometry (IOS) is a non-invasive method to assess lung function.
• Factors influencing variability in IOS measurements are not fully understood.
• Accurate reference equations are crucial for interpreting IOS data.

Purpose of the Study:

• To identify variables associated with IOS measurements in a well-screened population of active-duty service members (ADSMs).
• To develop and validate new predictive models for IOS measurements.
• To compare the performance of new models against existing reference equations.

Main Methods:

• Utilized IOS data from the STAMPEDE II cohort of predeployment ADSMs.
• Constructed predictive models incorporating demographic variables (age, height, weight, sex, race/ethnicity, military rank).
• Validated derived equations against postdeployment STAMPEDE II subjects and external ADSM IOS datasets.

Main Results:

• Demographic factors including age, height, weight, sex, race/ethnicity, and military rank were significantly associated with IOS measures (R5, R20, X5, fres, AX).
• Existing reference equations consistently underestimated IOS measurements in the STAMPEDE II cohort.
• Newly derived equations showed overestimation, while existing equations showed underestimation in external validation datasets.

Conclusions:

• IOS measurements in a well-screened ADSM population were higher than predicted by current reference equations.
• Demographic characteristics of the derivation populations, specifically race and military rank, contribute to discrepancies in predicted IOS values.
• Updated predictive models are needed to accurately interpret IOS measurements in ADSM populations.