Summarising published results from spirometric surveys of COPD: the problem of inconsistent definitions.

SETTING: The use of different spirometric definitions for chronic obstructive pulmonary disease (COPD) has made an informative review of the available prevalence surveys impossible. OBJECTIVE: To derive adjustment factors that allow the comparison of studies using different spirometric criteria. METHODS: Pre- and post-bronchodilator one-second forced expiratory volume (FEV1) and forced vital capacity (FVC) values were taken from the Burden of Obstructive Lung Disease (BOLD) survey in 16 centres. Using a post-bronchodilator FEV1/FVC ratio less than the lower limit of normal (LLN) as our reference prevalence, we calculated simple multiplicative adjustments to transform other reported prevalence estimates to reference values. These adjustments were then tested on independent data sets from six further BOLD centres and five centres from the PLATINO study, a Latin American survey on obstructive lung disease. RESULTS: Prevalence estimates based on pre-bronchodilator fixed-ratio measurements were 5-25% higher than reference values, and were strongly positively biased with age and prevalence level. Applying simple adjustments provided prevalence estimates that were almost unbiased and within 5% of the reference values. CONCLUSIONS: Using the BOLD data, we have been able to estimate COPD prevalences based on post-bronchodilator FEV1/FVC < LLN by adjusting estimates based on other common definitions, enabling more meaningful comparisons of published findings.