Background and objective: COPD is a complex condition with a heavy burden of disease. Many multidimensional tools have been studied for their prognostic utility but none has been universally adopted as each has its own limitations. We hypothesize that a multidimensional tool examining four domains, health-related quality of life, disease severity, systemic effects of disease and patient factors, would better categorize and prognosticate these patients.
Methods: We first evaluated 300 patients and found four factors that predicted mortality: BMI, airflow obstruction, St George's Respiratory Questionnaire and age (BOSA). A 10-point index (BOSA index) was constructed and prospectively validated in a cohort of 772 patients with all-cause mortality as the primary outcome. Patients were categorized into their respective BOSA quartile group based on their BOSA score. Multivariate survival analyses and receiver operator characteristic (ROC) curves were used to assess the BOSA index.
Results: Patients in BOSA Group 4 were at higher risk of death compared with their counterparts in Group 1 (hazard ratio (HR): 0.29, 95% CI: 0.16-0.51, P < 0.001) and Group 2 (HR: 0.53, 95% CI: 0.34-0.82, P = 0.005). Race and gender did not affect mortality. The area under the ROC curve for BOSA index was 0.690 ± 0.025 while that for Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2011 was 0.641 ± 0.025 (P = 0.17).
Conclusion: The BOSA index predicts mortality well and it has at least similar prognostic utility as GOLD 2011 in Asian patients. The BOSA index is a simple tool that does not require complex equipment or testing. It has the potential to be used widely.
Keywords: chronic obstructive pulmonary disease; classification; morbidity; mortality; prognosis.
© 2016 Asian Pacific Society of Respirology.