Factors Associated with Pertussis Testing: Potential Disparities in Case Ascertainment
Methods: We conducted a retrospective case-control study of children ≤18 years who presented to a primary care clinic within the Children’s Hospital of Philadelphia network between July 2011 and September 2013. Case patients (tested by pertussis PCR) were matched by date to control patients (seen for cough illness but not tested for pertussis). Logistic regression was used to compare children who were and were not tested for pertussis.
Results: Pertussis PCR was obtained from 6797 children (544 positive, 8.0%); 6236 untested controls were identified. On bivariate analysis, age group (p<0.001) and race (p<0.001) differed significantly by testing status, but sex, ethnicity, and insurance payor did not. Children tested for pertussis were less likely to have asthma (42.2% vs 57.8%, p<0.001) or documented fever (3.9% vs 5.7%, p<0.001), but were more likely to have ≥1 visits in the prior month (54.0% vs 41.1%, p<0.001) and to have been seen at an urban clinic site (36.2% vs 24.7%, p<0.001). Virtually all patients were up-to-date for pertussis vaccine (96%), and pertussis vaccine status did not differ between groups. Children tested for pertussis were more likely to have last received pertussis vaccine ≥3 years ago (33.5% vs 31.1%, p=0.004). On multivariate analysis, pertussis vaccine ≥3 years ago, ≥1 visit in the past month, and visiting an urban primary care center were associated with testing for pertussis (ORs [95% CI]: 1.19 [1.06-1.34], 1.67 [1.55-1.79], and 2.38 [2.16-2.63], respectively). Lower likelihood of pertussis testing was associated with age 1-6 years, black race, history of asthma, and fever (ORs [95% CI]: 0.83 [0.74-0.94], 0.55 [0.49-0.61], 0.71 [0.66-0.77], and 0.57 [0.48-0.68], respectively).
Conclusion: Pertussis testing practices varied by both patient and provider factors, with clinic site having the greatest effect. Racial disparities may also exist. Standardization of testing practices might improve pertussis case ascertainment.
D. Campos, None
K. Feemster, None