Evaluation of New and Repurposed Tools to Assess Post-Tuberculosis Lung Disease in Adolescents: A Cross-Sectional Analysis
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Tuberculosis, even when successfully treated, frequently leads to long-term sequelae, known as post-tuberculosis lung disease (PTLD). Adolescents account for over 1 million incident tuberculosis cases each year, and PTLD in this population may contribute greatly to the global burden of chronic lung disease. However, research to better understand and prevent adolescent PTLD is hampered by uncertainty regarding which tools best assess respiratory disability and lung function in this population.
Methods
In this cross-sectional analysis of 101 adolescent tuberculosis survivors in Lima, Peru, we administered the St. George’s Respiratory Questionnaire (SGRQ) to assess respiratory disability, and spirometry and oscillometry to measure lung function. We used factor analysis, correlations, and structural equation modeling to assess reliability and validity of an abbreviated SGRQ, oscillometry, and spirometry.
Results
Our abbreviated, 18-item SGRQ had high reliability (omega ≥0.90), internal structure validity (factor loadings for most questions >0.75), and external validity (correlation: -0.62 with overall health rating). More participants were able to complete oscillometry vs. spirometry (100% vs. 91.1%, p <0.0001). Oscillometry metrics had higher reliability (0.82-0.88) than spirometry metrics (0.62-0.76). SGRQ scores had small correlations with spirometry and minimal correlations with oscillometry. The combination of the SGRQ, oscillometry, and spirometry demonstrated good model fit as an overall assessment of lung health.
Conclusion
Our findings support the combination of an abbreviated version of the SGRQ, spirometry, and oscillometry for evaluating adolescent PTLD. These results pave the way for critical research to better understand the long-term impacts of tuberculosis on adolescent lung health.