In Reply: In response to Dr Linder, the difference in pretest probabilities for pertussis infection between those in studies we cited and those from the Acellular Pertussis Vaccine Trial (APERT)1 is probably due to differences in the study populations. The community-based studies we cited reflect the prevalence of pertussis infection among patients who sought care for persistent cough illness. In contrast, the APERT trial included cough illnesses that were reported by participants in response to bimonthly telephone calls from the study staff. It seems likely that the threshold for telephone reporting of persistent cough might be lower than the threshold for seeking medical care for the cough. Such a low threshold is supported by the large number of participants from both treatment and control groups who reported a cough illness lasting more than 3 weeks during the 2.5 years of the APERT study (1047/2781 [38%]). For the purpose of estimating the baseline probability of pertussis infection in a clinical care setting, we believe the estimate from a study population of patients who seek care is most appropriate. Importantly, the true prevalence of pertussis infection is difficult to determine because of the pathobiology of the infection and the limitations of available diagnostic tests.
Gonzales R, Cornia PB, Newman T. Diagnosis of Pertussis—Reply. JAMA. 2010;304(21):2358-2359. doi:10.1001/jama.2010.1726