BMC Infectious Diseases

Background The strikingly young age of new variant Creutzfeldt-Jacob disease (vCJD) cases remains unexplained. Age dependent susceptibility to infection has been put forward, but differential dietary exposure to contaminated food products in the UK population according to age and sex during the bovine spongiform encephalopathy (BSE) epidemic may provide a simpler explanation. Methods Using recently published estimates of dietary exposure in mathematical models of the epidemiology of the new variant Creutzfeldt Jacob disease (vCJD), we examine whether the age characteristics of vCJD cases may be reproduced. Results The susceptibility/exposure risk function has likely peaked in adolescents and was followed by a sharp decrease with age, evocative of the profile of exposure to bovine material consumption according to age. However, assuming that the risk of contamination was proportional to exposure, with no age dependent susceptibility, the model failed to reproduce the observed age characteristics of the vCJD cases: The predicted cumulated proportion of cases over 40 years was 48%, in strong disagreement with the observed 10%. Incorporating age dependent susceptibility led to a cumulated proportion of cases over 40 years old of 12%. Conclusions This analysis provides evidence that differential dietary exposure alone fails to explain the pattern of age in vCJD cases. Decreasing age related susceptibility is required to reproduce the characteristics of the age distribution of vCJD cases. Background Owing to the lack of data concerning age dependent dietary exposure to bovine material in food in the UK, mathematical models used in the study of new variant Creutzfeldt-Jacob disease (vCJD) have postulated an age risk function for contamination where the influence of age dependent dietary exposure to contaminated material could not be separated from intrinsic age related susceptibility [1-3]. For example, we modelled exposure/susceptibility by a plateau during the first 15 years of age followed by an exponential decrease afterwards [2]. Using this approach we estimated the duration of the incubation period for vCJD to circa 15 years; and predicted that the epidemic had likely peaked in 2001-2002 which is now consistent with the observed data [2, 3]. The simple representation of the age risk function implied that few parameters were necessary to describe it, a desirable property when estimation is based on limited data. By June 2003, 139 cases of vCJD had been reported to the vCJD surveillance unit in the UK, as compared to the 79 cases used in our original paper [2]. This warrants the inclusion of more detail in the age risk function, especially to investigate whether the constant risk assumption in children holds, as it led to predict a bimodal age distribution for cases in the coming years [3]. At the same time, detailed estimates of the dietary bovine material consumption in the UK population according to age and sex have been made available [4]. This offers the opportunity to try to disentangle the role of exposure from that of age-specific susceptibility. Indeed, the estimates for dietary exposure show that it peaked during adolescence and decreased with age afterwards, a pattern which is consistent with the finding that most vCJD cases are young and were therefore at most teenagers during the years when the bovine spongiform encephalopathy (BSE) epidemic was at its maximum. We therefore set out to estimate the age risk function with a versatile description based on step functions, and to investigate whether dietary exposure alone could explain the age distribution of cases.