© 2005 American Thoracic Society
Early Therapy in Chronic Obstructive Pulmonary DiseaseSelection Bias by Exclusion?From the Authors:We are grateful to Dr. Suissa for his comments, but unfortunately he is incorrect in saying that 33 exacerbations were untreated. As described in our article (1), 16 of these 33 exacerbations involved treatment before the formal onset of the exacerbation by our symptomatic definition and were excluded for this reason. Of the other 17 untreated exacerbations, the delay from onset to consultation (when treatment was either initiated or not) was 3.5 days (interquartile range 0–7.5), from onset to resolution of symptoms was 9 days (7–16), and from consultation to resolution was 5 days (1–11). These intervals were not significantly different (for all, p > 0.58) from the findings for the 625 treated exacerbations of 2 days (1–5), 9 days (5–15), and 6 days (2–12), respectively. Dr. Suissa's interpretation that these exacerbations were untreated because the patient had recovered from the exacerbation by the time that he or she was seen in the clinic is therefore incorrect. We encourage our patients to seek treatment, providing them with a dedicated mobile telephone number to make appointments and speak directly to a doctor, with the aim of seeing them within 48 hours. The reason these exacerbations were not treated was that, in the view of the consulting physician, they did not warrant treatment with oral antibiotics and/or oral steroids on clinical grounds. We have performed the analysis requested by Dr. Suissa and included these untreated exacerbations. The relationship between recovery and consultation delay was essentially unchanged (from 0.424 to 0.428 days/days delay) and remained highly significant (p < 0.001). Dr. Suissa's comments highlight the difficulties of using data on health care use, particularly for the definition of an exacerbation. In our studies, we have carefully and prospectively collected daily data and, using a well-validated definition, have been able to accurately determine the onset and resolution of an exacerbation. Furthermore, we have for the first time demonstrated that unreported, and hence untreated, exacerbations play an important role in determining both health-related quality of life and risks of hospitalization. We are grateful to Dr. Suissa, therefore, for raising this issue and allowing us to reemphasize the important clinical message of this study.
St. Bartholomew's Hospital London, United Kingdom FOOTNOTES Conflict of Interest Statement: G.C.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter; T.M.A.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter; J.R.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter; T.A.R.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter; J.A.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
