Open Access Research article

Human dose response relation for airborne exposure to Coxiella burnetii

Russell John Brooke1*, Mirjam EE Kretzschmar12, Nico T Mutters3 and Peter F Teunis24

Author Affiliations

1 Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, The Netherlands

2 Centre for Infectious Disease Control, RIVM, Bilthoven, The Netherlands

3 Department of Infectious Diseases, Medical Microbiology and Hygiene, Heidelberg University Hospital, Heidelberg, Germany

4 Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA, USA

For all author emails, please log on.

BMC Infectious Diseases 2013, 13:488  doi:10.1186/1471-2334-13-488

Published: 21 October 2013



The recent outbreak of Q fever in the Netherlands between 2007 and 2009 is the largest recorded Q fever outbreak. Exposure to Coxiella burnetii may cause Q fever but the size of the population exposed during the outbreak remained uncertain as little is known of the infectivity of this pathogen. The quantification of the infectiousness and the corresponding response is necessary for assessing the risk to the population.


A human challenge study was published in the 1950s but this study quantified the dose of C. burnetii in relative units. Data from a concurrent guinea pig challenge study were combined with a recent study in which guinea pigs were challenged with a similar aerosol route to quantify human exposure. Concentration estimates for C. burnetii are made jointly with estimates of the dose response parameters in a hierarchical Bayesian framework.


The dose for 50% infection (InfD50%) in human subjects is 1.18 bacteria (95% credible interval (CI) 0.76-40.2). The dose for 50% illness (IllD50) in challenged humans is 5.58 (95%CI 0.89-89.0) bacteria. The probability of a single viable C. burnetii causing infection in humans is 0.44 (95%CI 0.044-0.59) and for illness 0.12 (95%CI 0.0006-0.55).


To our knowledge this is the first human dose–response model for C. burnetii. The estimated dose response relation demonstrates high infectivity in humans. In many published papers the proportion of infected individuals developing illness is reported to be 40%. Our model shows that the proportion of symptomatic infections may vary with the exposure dose. This implies that presence of these bacteria in the environment, even in small numbers, poses a serious health risk to the population.

Q fever; Dose response; Challenge study; Risk analysis; Beta-poisson