Reply to Lin

Abstract

garding our article [2], and we appreciate the opportunity to respond. We agree that our wording was imprecise in the discussion of interpretation of the results of the ORs calculated in the multivariate analysis, and we agree that it is implausible that human Mycobacterium bovis tuberculosis (TB) is more likely than Mycobacterium tuberculosis TB, given that human M. bovis is much less prevalent than M. tuberculosis [1]. We interpreted the same statistics correctly and precisely in both the abstract and the results section of the document as: Patients who were not born in the United States, Hispanic patients, patients <15 years of age, patients reported to be HIV infected, and patients with extrapulmonary disease each had increased adjusted odds of having M. bovis versus M. tuberculosis TB [2, p. 168]. Moreover, the referent groups are clearly indicated in table 2. Our intention was to encourage US clinicians to consider M. bovis in their differential diagnosis, especially when patients have certain demographic and clinical characteristics. We are confident that the data reported in the tables and in the remainder of the document provide the evidence supporting this intention. We are grateful for this opportunity to underscore the importance of precision when communicating interpretations of the results of multivariate analysis [3, 4], of paying particular attention to distinguishing between the ORs and odds, and of indicating whether they are adjusted and what factors were controlled for when