Assessment of the internal structure of GOLD 2011 system

Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2011 introduced a new multidimensional system (symptom/risk) for assessment chronic obstructive pulmonary disease (COPD). The aim of this study was to explore the construct validity of the GOLD 2011 classification strategy; specifically, we evaluated its internal structure in terms of reliability and exploratory factor analysis (EFA).

Reliability (Cronbach alpha coefficient), correlations between variables and two successive EFA were performed to assess the internal structure of GOLD 2011. Symptoms (mMRC dyspnea score) and risk (number of previous year exacerbations and forced expiratory volume in the first second % of predicted) were selected as variables.

The analysis included 679 COPD patients from two Spanish cohorts (71.4 ± 11.7 years). Alpha coefficient of the 3 items was 0.52 for the whole sample. Variables presented statistically significant correlations, but of low to moderate magnitude. A first EFA extracted only one factor accounting 52% of the total variance. A second EFA including four items (the three GOLD 2011 variables plus comorbidities as Charlson index score), extracted two-factors accounting for 65% of the total variance. The first factor included the three items of GOLD 2011, and the second contained only one variable (comorbidities). This solution was stable in patients with different levels of COPD severity.

The available evidence suggests that GOLD 2011 strategy presented a low reliability, and its theorized multidimensionality was not confirm by EFA. Comorbidities appears as a separate and independent domain.