to the editor: Drs. Stephens and Yew provided a well-written and comprehensive overview of the diagnosis of chronic obstructive pulmonary disease (COPD). They appropriately referenced the National Heart, Lung, and Blood Institute (NHLBI) recommendation of spirometry for suspected COPD,¹ and they indicated in a table that this recommendation has a Strength of Recommendation Taxonomy (SORT) C rating. Family physicians may assume the NHLBI recommendation published in 2000¹ is the best evidence available and applicable to their practice. However, of the 23 executive committee members of the National Lung Health Education Program (the source for the NHLBI recommendation), only one is a family physician, and the preponderance are chest physicians.²

A 2005 systematic review on the use of spirometry for case finding, diagnosis, and management of COPD commissioned by the Agency for Healthcare Research and Quality had different findings.³ The authors found that: (1) clinical improvement was not associated with a patient's spirometric response to therapy; (2) treatments other than smoking cessation did not alter the rate of spirometric decline over time; (3) there was wide intra-individual variation in spirometric decline; and (4) interventions were not effective in asymptomatic persons or those with mild to moderate airflow obstruction. The primary benefit of spirometry is to identify persons who might benefit from pharmacologic treatment to improve exacerbations (e.g., adults with symptomatic, severe to very severe airflow obstruction). Following the NHLBI guideline, they estimated that 941 patients would need to be screened to reduce one person's COPD exacerbations.³

The authors of the systematic review concluded that the evidence does not support widespread use of spirometry in primary care settings for all adults with persistent respiratory symptoms or those with a history of exposure to pulmonary risk factors for case finding, improving smoking cessation rates, monitoring the clinical course of COPD, or adjusting COPD interventions; that routine spirometric testing in primary care settings is likely to result in considerable testing and treatment costs, resource use, and health care personnel time; and that it is likely to label a large number of persons (many not reporting bothersome respiratory symptoms or having nondisabling symptoms) as diseased who would not benefit from testing or treatment.³ The American College of Physicians' guideline on the diagnosis and management of stable COPD is consistent with this evidence.⁴

Although American Family Physician articles routinely include a SORT ratings table, readers' understanding of the ratings is important for appropriate translation into practice. Remember that a C rating reflects “consensus, disease-oriented evidence, usual practice, expert opinion, or case series.”⁵ Readers should be made aware of the best available evidence and consider the level of evidence when deciding to make any patient-oriented practice changes as a result.

in reply: We are grateful for Dr. Fink's thoughtful reply to our article on diagnosing chronic obstructive pulmonary disease (COPD). Dr. Fink draws particular attention to two key issues: the difference in the role of spirometry for diagnosing and screening for COPD; and the need for caution against the indiscriminate application of guideline recommendations. This is apparent in reviewing the final Strength of Recommendation Taxonomy (SORT) recommendation from our article that “spirometry should be performed in patients 45 years or older who smoke and have a persistent cough.” The citation on which it is based¹ qualifies it as a SORT C recommendation (consensus, disease-oriented evidence, usual practice, expert opinion, or case series). Other evidence, as presented by Dr. Fink, suggests that spirometry for case finding is not supported by current evidence.² In this context, spirometry is unlikely to be beneficial unless “future studies establish that spirometry improves smoking cessation rates, treatments other than smoking cessation benefit individuals with airflow obstruction who do not report respiratory symptoms, or that relative effectiveness between therapies varies according to an individual's baseline or follow-up spirometry.”² We do not suggest that spirometry should be used as a random screening tool. This same review by the Agency for Healthcare Research and Quality affirms that spirometry can augment the clinical examination and improve the accuracy of the diagnosis of COPD in patients whom it is clinically suspected.²

Dr. Fink's letter appropriately highlights the need to carefully and critically examine clinical guidelines before applying them in clinical practice. This includes understanding the composition of expert panels, potential subspecialty bias, and clinical relevance as it relates to community-based family medicine and primary care.

editor's note: The U.S. Preventive Services Task Force (USPSTF) recommends against screening for COPD using spirometry in adults who do not recognize or report respiratory symptoms to a physician, regardless of smoking status. The USPSTF found fair evidence that providing smokers with spirometry results does not independently improve cessation rates beyond those achieved using proven counseling and pharmacologic interventions.¹

KENNY LIN, MD

Associate Editor