Incidence of pertussis infection in healthcare workers

OBJECTIVE: To determine the incidence of pertussis infection in two groups of healthcare workers. DESIGN: Retrospective cohort study. SETTING: 660-bed, urban, tertiary-care university hospital. PARTICIPANTS: 106 resident physicians and 39 emergency department employees . INTERVENTIONS: Antibodies to pertussis toxin and filamentous hemagglutinin were determined in fresh serum specimens and in stored sera collected 1 to 3 years previously. A 50% rise in both the pertussis toxin and filamentous hemagglutinin from the initial to the follow-up specimen was considered dia gnostic of a pertussis infection. RESULTS: Two of 106 residents had serological evidence of a pertussis infec tion during 151.3 subject-observation years, for an annual incidence rate o f 1.3% (95% confidence interval [CI95], 0%-3.5%). Three of 39 emergency dep artment employees had serological evidence of a pertussis infection during 81.2 subject-observation years, for an annual incidence of 3.6% (CI95, 0%-9 .6%). Of these 5 subjects, 2 had symptomatic disease. CONCLUSION: We found both symptomatic and asymptomatic pertussis infections in two cohorts of healthcare workers. Although the incidence rates were so mewhat lower than found in other studies, they nonetheless were higher than for almost all other diseases for which we vaccinate healthcare workers. O ur results would support the use of acellular pertussis vaccine in healthca re workers (Infect Control Hasp Epidemiol 1999;20:120-123).