International Journal of Epidemiology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (15) Request Permissions Google Scholar Articles by Cooper, J. Articles by Bird, S. Search for Related Content PubMed PubMed Citation Articles by Cooper, J. Articles by Bird, S. Social Bookmarking          What's this?

International Journal of Epidemiology 2003;32:784-791
© International Epidemiological Association 2003

Special Theme: Infectious Diseases

Predicting incidence of variant Creutzfeldt-Jakob disease from UK dietary exposure to bovine spongiform encephalopathy for the 1940 to 1969 and post-1969 birth cohorts

JD Cooper¹ and SM Bird²

¹ MRC Biostatistics Unit and Department of Medical Genetics, Cambridge.
² Medical Research Council Biostatistics Unit, Institute of Public Health, Cambridge, UK.

SM Bird, Medical Research Council Biostatistics Unit, Institute of Public Health, Cambridge, UK. E-mail: sheila.bird{at}mrc-bsu.cam.ac.uk

	Abstract

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
Background To investigate variant Creutzfeldt-Jakob disease (vCJD) incubation period, transmission barrier, and short-term vCJD predictions for methionine homozygotes in 1940–1969 and post-1969 birth cohorts by use of gender- and age-specific exposure intensities to bovine spongiform encephalopathy (BSE), based on consumption of beef mechanically recovered meat (MRM) and head meat.

Methods Simulation (from vCJD infections generated randomly from gender and age-specific dietary exposure intensities to BSE), constrained to equal the 47 and 64 vCJD onsets before 2001 in 1940–1969 and post-1969 birth cohorts, was used to estimate lognormal (and other) incubation mean and standard deviation which fitted the calendar year distribution of observed vCJD onsets; and to explore exponential decay in susceptibility to infection with age above 15 years.

Results For the post-1969 birth cohort, the best-fitting lognormal incubation period mean of 11 years (SD 1.5 years and 195 infections) was associated with 194 vCJD onsets (64 before 2001, 105 in 2001–2005, and 25 in 2006–2010). About one-fifth of simulated vCJD onsets before 2001 arose from infections in 1990–1996; age and gender of simulated and observed vCJD patients agreed closely. For the 1940–1969 birth cohort, well-fitting lognormal means ranged widely, the marginally best fitting being 26 years (SD 16.5 years and 382 infections; 47 vCJD onsets before 2001, 58 in 2001–2005, and 63 in 2006–2010). An age-dependent susceptibility function was required to match the age distribution of vCJD patients in the 1940–1969 birth cohort.

Conclusions About three-fifths of predicted vCJD onsets are expected to be in males, and nearly two-thirds of vCJD onsets in 2001–2005 are expected to be in post-1969 birth cohort according to best-fitting predictions.

---

Keywords vCJD, incubation period, predictions, dietary exposure intensities to BSE, birth cohort, simulation

Accepted 23 May 2003

Dietary exposure to the agent responsible for bovine spongiform encephalopathy (BSE) is the most likely primary cause of variant Creutzfeldt-Jakob disease (vCJD). Burgers, sausages, and other meat products have been implicated in human exposure to BSE through their use of beef mechanically recovered meat (MRM)^1–5 and head meat.⁶ Beef MRM and head meat were potentially contaminated respectively with BSE infected spinal cord and dorsal root ganglia, and brain. Legislation ended UK production of beef MRM in December 1995 and of head meat in March 1996.

Published vCJD predictions have differed in five important aspects, the first of which is the differences in the assumed shape of human exposure intensity to BSE: specifically, whether^7–9 and how^10–12 infectivity from clinical BSE and pre-clinical bovines was taken into consideration. Second are differences in the assumed impact of legislation, such as the 1989 Specified Bovine Offal (SBO) legislation, to prevent BSE infectivity from reaching consumers. The SBO legislation has commonly been assumed to have had a key role in reducing human exposure to BSE.^7–9 Third, a range of unimodal vCJD incubation period distributions has been assumed. Typically, they have included the lognormal distribution,^7,9,12 sometimes as a special case of a more flexible distribution.^8,10 Fourth is variation in whether any plausibility constraints have been imposed on vCJD incubation period mean, such as less than 30 years,^7,11,12 or on variation about a given mean.^7,12 Whether, and how, age-dependencies, such as in exposure,^10,11 susceptibility,^8,9 and incubation,^10,11 were introduced is the fifth aspect. In the absence of age-specific exposure intensities to BSE, age-dependencies cannot be distinguished.

Previously,^4–6 we have estimated dietary exposure intensities to BSE through consumption of beef MRM and head meat in burgers, sausages, and other meat products by birth cohort (pre-1940, 1940–1969, and post-1969) and gender; and highlighted major sensitivity issues. We now use these directly estimated dietary exposure intensities to make short-term predictions of vCJD incidence in the 1940–1969 and post-1969 birth cohorts. No predictions are made for the pre-1940 birth cohort as there has only been one vCJD onset reported in a patient born before 1940. As all vCJD patients to date are of one particular genotype (homozygous for methionine at polymorphic residue 129 of the prion protein—which accounts for about 40% of the UK population¹³), short-term predictions of vCJD incidence apply only to this subpopulation.^7–12

In addition to the dietary exposure intensities, our approach differs from previous models in: (1) we allow the incubation period distribution to differ between birth cohorts; (2) consider age-dependent susceptibility, based on the age-risk function of Valleron et al.;⁹ and (3) estimate the transmission barrier for individuals with the affected genotype.

	Methods

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
The number and calendar time of observed vCJD onsets depend on three factors: (1) how many individuals were infected; (2) calendar time of their infection; and (3) their vCJD incubation period, all unknown. To estimate these unknowns and to make short-term predictions of vCJD incidence, we adopted a simulation approach. Calendar times of infection are estimated from our previously estimated dietary exposure intensities to BSE,^4–6 which are shown in Figure 1 as proportion of cohort-specific exposures. Infectivity is expressed as bovine oral ID₅₀ (Bo ID₅₀) units, the oral dose required to cause infection in 50% of an exposed bovine population. Figure 1 shows two infectivity options. Option 1 assumes that BSE-infected bovines slaughtered in the 12 months before BSE onset are 54% as infectious as clinical BSE bovines, based on an assumed exponential progression of infectivity in the last year of the incubation period with a doubling time of 6 months.^5,6 Other doubling times (2 or 4 months) have been assumed as there is great uncertainty about the progression of infectivity, for example, in the forthcoming work on the Over Thirty Months Slaughter scheme in the UK.¹⁴ Option 2 (worse case) assumes that these BSE-infected bovines are as infectious as clinical BSE bovines.^5,6

View larger version (20K): [in this window] [in a new window]   Figure 1 Dietary exposure intensities to the bovine spongiform encephalopathy (BSE) agent through the consumption of beef mechanically recovered meat (MRM) and head meat in burgers, sausages, and other meat products)

 
The 1940–1969 birth cohort was the largest birth cohort (about 24.5 million people) and most exposed to BSE in beef MRM and head meat. It consumed 325 500 (560 500 for infectivity option 2) Bo ID₅₀ units with the post-1969 cohort in second place at 271 350 (457 700) Bo ID₅₀ units. The post-1969 birth cohort ranged from about 8.2 to 20.5 million people between 1980 and 1996. The pre-1940 birth cohort, included for completeness, was the least exposed, having consumed 86 500 (138 000) Bo ID₅₀ units; population ranged from about 23.4 to 13.5 million people. The distinct bimodality of the dietary exposure intensities to BSE primarily resulted from the assumed full reporting of BSE suspects after June 1988^5,6 and the removal of confirmed clinical BSE bovines from the human food chain in August 1988. The drop in demand for beef MRM in the late 1980s also contributed to the bimodal shape⁵ (for further discussion and sensitivity of assumptions see refs 5,6).

	Simulation of vCJD incidence

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
We simulated vCJD incidence within a birth cohort by generating the calendar times of infection, incubation periods (and hence, calendar times of onset), and survival to onset for n individuals, where n is the number of infections required to reproduce exactly the total number of observed onsets before 2001—the only model constraint.

To estimate the calendar years of infection for the n infected individuals, we made the strong assumption that vCJD infections predominantly resulted from exposure to BSE in beef MRM and head meat, and that we had successfully modelled this exposure.^4–6 Consequently, we generated the calendar times of infection for the n infected individuals from their birth cohort’s dietary exposure intensity to BSE by assuming that the number of infections in calendar year y (during exposure period 1980–1996) was proportional to the birth cohort’s dietary exposure to BSE in year y. For each infected individual, we then randomly assigned a week within their calendar year of infection when the event occurred. The age at infection (and hence year of birth) and gender of an infected individual were estimated in accordance with the age- and gender-specific exposure within the birth cohort in calendar year y. Infectivity option 1 was assumed for the main simulations, but simulations were repeated for infectivity option 2.

To estimate the incubation periods for the n infected individuals, we assumed a lognormal distribution^7,9,12,15 and searched for the best-fitting standard deviation for a given mean (see Incubation period distribution, below). Simulations were repeated assuming gamma and Weibull distributions for the incubation period.^7,16 For each infected individual, after generating their incubation period, we derived their calendar time of onset and age at onset. Finally, we used age- and gender-specific UK life tables¹⁷ to estimate whether the infected individuals survived to vCJD onset.

	Incubation period distribution

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
We searched through incubation period parameter combinations (mean and standard deviation) by simulating one hundred ‘epidemics’ for each combination and evaluating the consistency of expected vCJD simulated onsets with the observed onsets using an approximate ² measure of goodness-of-fit (see Goodness-of-fit criterion, below). The parameter search consisted of fixing the mean while epidemics for a range of standard deviations (increasing in 0.5-year steps) were evaluated. For incubation means from 10 to 30 years in integer steps, the parameter search identified the standard deviation that combined with it to produce the best (that is: lowest) goodness-of-fit test statistic.

The chosen range of incubation means was guided by the incubation means for other human transmissible spongiform encephalopathies (TSE).^16,18,19 In the absence of a species barrier, Collinge¹⁹ suggested an incubation mean of about 10–15 years (range: at least 4–40 years) for human TSE after peripheral innoculation or oral exposure, which may result in a longer incubation mean.

	Goodness-of-fit criterion

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
Each simulation was required to reproduce the total number of observed onsets within a given birth cohort before 2001, which were reported by the end of 2001, thereby reducing the impact of reporting delays. There had been 47 onsets in the 1940–1969 birth cohort and 64 onsets in the post-1969 birth cohort.

Expected vCJD simulated onsets were estimated by the mean number of simulated onsets for the 100 simulation runs (note: Tables give median). To compare the expected vCJD onsets with those observed before 1996, in 1996, 1997, 1998, 1999, and 2000, we used a standard ² measure of goodness-of-fit on 5 degrees of freedom (d.f.). If the goodness-of-fit test statistic exceeded 11.07, the 95% critical value for the ²₅ distribution, the simulated expectations were not consistent with the observed incidence. For the combined 1940–1969 and post-1969 birth cohorts, the simulation was required to reproduce the 111 onsets in the combined birth cohorts (the model constraint) and the goodness-of-fit test statistic on 11 d.f. measured agreement by birth cohort between expected vCJD simulated onsets and those observed before 1996, and in 1996, 1997, 1998, 1999, and 2000.

	Transmission barrier

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
Under the simplistic assumption that all exposed individuals with the same genotype as vCJD patients become infected, a bovine-to-human ‘transmission barrier’ for this subpopulation of a birth cohort can be approximated as their dietary exposure to BSE divided by the simulated number of infected individuals.

transmission barrier =total dietary exposure to BSE*0.4

n

However, such an estimate depends to a great extent on the estimates of bovine tissue-specific infectivity used in the estimation of the dietary exposure intensities, about which there is uncertainty.^20–22 The route of infection is also critical: the oral route of infection is known to be the least effective.²² Despite the hazards of interpretation, we estimated the transmission barrier for each simulation model to investigate whether the transmission barrier is similar for the best-fitting models of the 1940–1969 and post-1969 birth cohorts. As the above definition of the transmission barrier does not take into account the infective dose or the number of individuals exposed who do not become infected (not estimated by the model), the transmission barrier is different from the species barrier.¹⁹

	Age-dependent susceptibility

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
To improve the age-related characteristics of simulated patients in the 1940–1969 birth cohort, we included a modified version of the age-dependent susceptibility function used by Valleron et al.⁹ We assumed that all exposed individuals aged less than 15 years become infected but that the infection risk decreases exponentially with age thereafter. The rate of the exponential decay can be estimated within the simulation by introducing an age-related model constraint. As we have already estimated the incubation period parameters, we simply searched for the rate of exponential decay that best reproduced the observed patients’ period of birth. To compare the expected number of simulated patients born in 1940–1949, 1950–1959, 1960–1964, and 1965–1969 with those observed, we used a ² goodness-of-fit test statistic on 3 d.f.

	Results

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
Simulated vCJD onsets from the models for the combined 1940– 1969 and post-1969 birth cohorts, which assumed a common incubation period distribution and transmission barrier, were not consistent with the 111 observed vCJD onsets. As expected, given the higher dietary exposure but lower incidence in the 1940–1969 birth cohort, an excess of simulated onsets occurred in the 1940–1969 birth cohort and, consequently, too few in the post-1969 birth cohort.

For the 1940–1969 and post-1969 birth cohorts separately, Table 1 lists the best fitting standard deviation for lognormal incubation means of 10–30 years and associated ² goodness-of-fit test statistic. The ² test statistic was not significant for any of the models, indicating that their expected simulated vCJD onsets were consistent with the observed onsets. The ² test statistic had a distinct minimum value for the post-1969 birth cohort at an incubation period of 11 years (SD 1.5 years). For the 1940–1969 birth cohort, the ² test statistic was relatively flat around the minimum value at an incubation mean of 26 years (SD 16.5 years) and remained at about this level even with an incubation mean of 60 years.

View this table: [in this window] [in a new window]   Table 1 Best fitting models assuming a lognormal incubation period with means of 10–30 years

 
Predictions for the best fitting models for the 1940–1969 and post-1969 birth cohorts are summarized in Table 2. In the short-term, low vCJD incidences are predicted in both the 1940–1969 and post-1969 birth cohorts, respectively, 58 (5th–95th percentiles: 42–79) and 105 (85–140) onsets in 2001–2005 and 63 (45–86) and 25 (17–35) onsets in 2006–2010. Consequently, for the post-1969 birth cohort, primary vCJD incidence is predicted to peak in 2001–2005 with very few onsets after 2010. In contrast, for the 1940–1969 birth cohort, just under half the onsets are predicted to occur after 2010. The expected transmission barrier for the post-1969 birth cohort (558; 5th–95th percentiles: 462–646) was 1.5 times that for the 1940–1969 birth cohort (369: 292–476). The 1940–1969 birth cohort with incubation mean of 22 years had a similar transmission barrier to the post-1969 birth cohort (Table 1). Importantly, the simulated patient characteristics (gender and age [period of birth]), which were not used in the model fit, are strikingly similar to those of the observed patients in the post-1969 birth cohort. Older age at onset of the simulated patients in the 1940–1969 birth cohort was as a result of too few born in 1965–1969 (Table 3). Incorporating exponential decay in susceptibility at a rate of 0.06 per year (95% CI: 0.02, 0.11) after 15 years of age reduced simulated patients’ expected age at onset to 35.5 years (5th–95th percentiles: 34.7–37.4 years) and, accordingly, more simulated patients were born in 1965–1969. The median number of exposures was then 828 (5th to 95th percentiles: 679–1009), which resulted in 332 infections (267–410), of whom 43 (30–56) died before onset.

View this table: [in this window] [in a new window]   Table 2 Variant Creutzfeldt-Jakob disease (vCJD) predictions with best-fitting lognormal incubation period

 

View this table: [in this window] [in a new window]   Table 3 Concordant age distributions: best-fitting log normal variant Creutzfeldt-Jakob disease (vCJD) predictions without and with exponential decay in susceptibility with age

 
According to the best-fitting models, about 23% of the vCJD patients in the 1940–1969 birth cohort and 19% of the patients in the post-1969 birth cohort with vCJD onset before 2001 were infected in 1990–1996. Uncertainty about expected vCJD infections in the 1940–1969 and post-1969 birth cohorts respectively ranged from 80 to 445 and 155 to 1985. To illustrate the range of lognormal models for the 1940–1969 birth cohort which have a similar fit, selected output from models with incubation means of 11 (mean of best fitting model for post-1969 birth cohort), 22 (closest transmission barrier to post-1969 birth cohort), and 35 (example of low ² test statistic for log normal means up to 60 years) years are summarized in Table 4. There was little difference between the models in the characteristics of simulated patients with onset before 2001. As expected, the longer the incubation period, the more infections were generated, the greater the number of infected individuals who die from other causes before vCJD onset, and the lower the transmission barrier. Low predictions of vCJD incidence persist even with an incubation mean of 60 years (data not shown).

View this table: [in this window] [in a new window]   Table 4 Variant Creutzfeldt-Jakob (vCJD) predictions for 1940 to 1969 birth cohort using selected log normal incubation periods

 
Best fitting models assuming gamma and Weibull incubation period distributions had identical values for the incubation period parameters as the lognormal incubation period distribution for the post-1969 birth cohort and incubation means of 28 years (SD 14 years) and 30 years (SD 12 years) respectively for the 1940–1969 birth cohort. The ² test statistic was again not significant for any of the models and relatively flat for the 1940–1969 birth cohort models.

For infectivity option 2, the best-fitting models assuming a lognormal incubation period distribution had identical incubation period parameters as infectivity option 1 for the post-1969 birth cohort and an incubation mean of 16 years (SD 9 years) for the 1940–1969 birth cohort. The ² test statistic was again not significant for any of the models and relatively flat for the 1940–1969 birth cohort models. Expected transmission barrier for the post-1969 birth cohort (770; 5th–95th percentiles: 652–902) was 0.6 times lower than that of the 1940–1969 birth cohort (1310; 5th–95th percentiles: 1120–1578). With similar or shorter incubation means, the birth cohort specific transmission barriers were necessarily higher for infectivity option 2 than for infectivity option 1.

As summarized in Table 5, when incubation mean ranged from 10 to 30 years, simulated vCJD infections ranged from 80 to 1140 for the 1940–1969 birth cohort and from 155 to 6585 for the post-1969 birth cohort, dependent on assumed incubation distribution and infectivity option. The transmission barrier ranged from 15 to over 2500. However, constraints on incubation mean (maximum of 20 years) or transmission barrier (between 201 and 800) dramatically reduced the prediction range for both birth cohorts.

View this table: [in this window] [in a new window]   Table 5 How constraints affect simulated median number of infections (transmission barrier)

 

	Discussion

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
Simulation models for the combined 1940–1969 and post-1969 birth cohorts revealed that if we assume a common incubation period and transmission barrier, given the higher exposure but lower incidence in the 1940–1969 birth cohort, an excess of simulated onsets occurred in the 1940–1969 birth cohort and too few in the post-1969 birth cohort. To reduce the number of onsets in the 1940–1969 birth cohort and increase the number in the post-1969 birth cohort, the model would require some kind of age dependence in incubation or susceptibility.

Separate simulation models for the 1940–1969 and post-1969 birth cohorts suggested that the incubation period was shorter for the post-1969 birth cohort. However, there was greater uncertainty about the incubation period for the 1940–1969 birth cohort. Previous work has reported a very severe incubation period parameter identifiability problem.⁸

The age-dependent susceptibility function was required to match the age distribution of vCJD patients in the 1940–1969 birth cohort. The simple formulation proposed by Valleron et al.⁹ of exponential decay in susceptibility after 15 years of age worked well.

Our simulation models make low predictions of vCJD incidence in methionine homozygotes. Short-term predictions from the best fitting models for the 1940–1969 and post-1969 birth cohorts respectively were 58 (5th–95th percentiles: 42–79) and 105 (5th–95th percentiles: 85–140) onsets in 2001–2005. In the short-term, the predicted low incidence will make it hard to distinguish between incubation periods. For instance, expected vCJD incidence in 2001–2005 for the 1940–1969 birth cohort ranged only between 36 and 52 for incubation means ranging from 11 to 35 years.

We have assumed that the primary sources of exposure to BSE were through the consumption of meat products containing beef MRM and head meat. Other global and local sources of exposure to BSE have not been ruled out, but may be unquantifiable. The cluster of five vCJD patients from Queniborough in Leicestershire raised the spectre of localized sources of exposure to BSE. The investigation into the cluster revealed that traditional practices of local butchers were likely to have resulted in the contamination of bovine carcass meat with BSE.²³ Had such localized exposures occurred throughout the UK the cluster of patients from Queniborough would have stood out less from the other vCJD patients. Predictions of vCJD incidence based on global exposure intensities to BSE need to consider that vCJD patients from Queniborough could have succumbed to local exposure. Practically, this would suggest the exclusion of vCJD patients from Queniborough from those for whom any global exposure intensity to BSE was responsible. The exclusion of the vCJD patients from Queniborough did not change our results.

More recent findings of the BSE Pathogenesis Experiment have resulted in a shift from 10 Bo ID₅₀ units,²⁰ as assumed in our estimation of the dietary exposure intensities to BSE,^5,6 to 50 Bo ID₅₀ units per gram of brain, spinal cord, and dorsal root ganglia of a clinical BSE bovine.¹⁴ This has only a multiplicative effect on our dietary exposure intensities to BSE.

The strength of the method used in this paper is that calendar times of infection and infected patient characteristics are derived directly from dietary exposure intensities to BSE. We used a transparent simulation model with a single constraint and, for each birth cohort, had only to estimate the two incubation period parameters and the exponential decay in the age-dependent susceptibility function. Clearly, the simulation model can be extended to include more flexible incubation period distributions and different age-dependent susceptibility functions.

Considerable uncertainty about the incubation period will remain for a number of years, particularly for the 1940–1969 birth cohort. Despite this incubation period uncertainty, overall low predictions of vCJD incidence within the subpopulation of the UK with the affected genotype are made. According to the best-fitting models, almost two-thirds of predicted vCJD patients in 2001–2005 are expected to be in the post-1969 birth cohort, whose vCJD incidence is predicted to peak in this calendar period and to outnumber (about 1.5 times more onsets) those in the 1940–1969 birth cohort. Very few onsets are predicted to occur after 2010 in the post-1969 birth cohort. In contrast, for the 1940–1969 birth cohort, almost half of onsets are predicted to occur after 2010. About three-fifths of vCJD patients are expected to be male.

The dietary exposure intensities to BSE show that the age distribution of observed vCJD patients can only arise if younger individuals have a shorter incubation period and/or are more susceptible to infection. There is remarkable similarity between the age distribution and gender of simulated and observed vCJD patients, which supports (but does not prove) our assumption about the primary sources of exposure to BSE. In contrast to other predictions of vCJD incidence, our predictions are testable by the number, birth cohort, age, and gender of future vCJD patients.

As all vCJD patients tested have been homozygous for methionine at polymorphic residue 129 of the prion protein, it is premature to make extrapolations from vCJD incidence in individuals with the affected genotype to individuals with other genotypes. A number of biological processes, including species barrier¹⁹ and incubation period,⁸ may be dependent on genotype. For instance, other genotypes may be absolutely protected, relatively less susceptible, or equally susceptible but with substantially longer incubation periods. Transmission studies, however, argue against absolute protection.²⁴

	Acknowledgements

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
vCJD data were kindly provided by the National Creutzfeldt-Jakob Disease Surveillance Unit. Funding was from the Medical Research Council and the Department for Environment, Food, and Rural Affairs.

KEY MESSAGE Directly estimated dietary exposure intensities to bovine spongiform encephalopathy (BSE) suggest that younger individuals are more susceptible to BSE agent.

 

	References

Top Abstract Methods Simulation of vCJD incidence Incubation period distribution Goodness-of-fit criterion Transmission barrier Age-dependent susceptibility Results Discussion Acknowledgements References

 
¹ Gore SM, Bingham S, Day NE. Age related dietary exposure to meat products from British dietary surveys of teenagers and adults in the 1980s and 1990s. BMJ 1997;315:404–05.[Free Full Text]

² CJD Support Network. CJD and Prion Disease: An Introduction and Explanation. Alzheimer’s Disease Society; 1998.

³ Phillips Lord, Bridgeman J, Ferguson-Smith M. The BSE Inquiry, Vol. 6: Human Health, 1989–96. London: The Stationery Office; 2000. URL: http://www.bse.org.uk/volume6/toc.htm

⁴ Cooper JD, Bird SM. UK bovine carcass meat consumed as burgers, sausages and other meat products: by birth-cohort and gender. J Cancer Epidemiol Prevent 2002;7:49–57.

⁵ Cooper JD, Bird SM. UK dietary exposure to BSE in beef mechanically recovered meat: by birth-cohort and gender. J Cancer Epidemiol Prevent 2002;7:59–70.

⁶ Cooper JD, Bird SM. UK dietary exposure to BSE in head meat: by birth-cohort and gender. J Cancer Epidemiol Prevent 2002;7:71–83.

⁷ Cousens SN, Vynnycky E, Zeidler M, Will RG, Smith PG. Predicting the CJD epidemic in humans. Nature 1997;385:197–98.[CrossRef][Medline]

⁸ Huillard D’Aignaux JN, Cousens SN, Smith PG. Predictability of the UK variant Creutzfeldt-Jakob disease epidemic. Science 2001;294: 1729–31.[Abstract/Free Full Text]

⁹ Valleron AJ, Boelle PY, Will R, Cresbon JY. Estimation of epidemic size and incubation time based on age characteristics of vCJD in the United Kingdom. Science 2001;294:1726–28.[Abstract/Free Full Text]

¹⁰ Ghani AC, Ferguson NM, Donnelly CA, Hagenaars TJ, Anderson RM. Epidemiological determinants of the pattern and magnitude of the vCJD epidemic in Great Britain. Proc R Soc Lond Biol 1998;265:2443–53.[Medline]

¹¹ Ghani AC, Ferguson NM, Donnelly CA, Anderson RM. Predicted vCJD mortality in Great Britain. Nature 2000;406:583–84.[CrossRef][Medline]

¹² Cooper JD, Bird SM, De Angelis D. Prevalence of detectble abnormal prion protein in persons incubating vCJD: plausible incubation periods and cautious inference. J Epidemiol Biostat 2000;5:209–19.[Medline]

¹³ Collinge J, Palmer MS, Dryden AJ. Genetic predisposition to iatrogenic Creutzfeldt-Jakob disease. Lancet 1991;337:1441–42.[CrossRef][Web of Science][Medline]

¹⁴ Comer PJ, Huntly PJ. TSE risk assessment—a decision support tool. Stat Meth Med Research 2003;12:279–91.

¹⁵ Sartwell PE. The distribution of incubation periods of infectious disease. Am J Hyg 1950;51:310–18.[Web of Science][Medline]

¹⁶ Huillard d’Aignaux JN, Cousens SN, Maccario J et al. The incubation period of kuru. Epidemiology 2002;13:402–08.[CrossRef][Web of Science][Medline]

¹⁷ Government Actuary’s Department. URL: http://ww.gad.gov.uk/

¹⁸ Brown P, Preece M, Brandel J et al. Iatrogenic Creutzfeldt-Jakob disease at the millennium. Neurology 2000;55:1075–81.[Abstract/Free Full Text]

¹⁹ Collinge J. Variant Creutzfeldt-Jakob disease. Lancet 1999;354: 317–23.[CrossRef][Web of Science][Medline]

²⁰ Comer PJ, Spouge J. Application of Risk Assessment to Estimating Human Risks. from BSE [Internal report]. Det Norske Veritas 1998.

²¹ Diringer H. Bovine spongiform encephalopathy (BSE) and public health. In: Aggett PJ, Kuiper HA (eds). Risk Assessment in the Food Chain of Children. Nestlé Nutrition Workshop Series;44:225–33. Philidelphia: Vevey/Lippincott Williams and Wilkins Publishers 1999.

²² Oral exposure of humans to the BSE agent: infective dose and species barrier. Scientific Steering Committee of European Union opinion. URL: http://europa.eu.int/comm/food/fs/sc/ssc/

²³ Ashraf H. UK investigators put forward theory for vCJD cluster. Lancet 2001;357:937.[Medline]

²⁴ Raymond GJ, Hope J, Kocisko DA et al. Molecular assessment of the potential transmissibilities of BSE and scrapie to humans. Nature 1997;388:285–88.[CrossRef][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				C. M. Ott, A. Akhavan, and V. R. Lingappa Specific Features of the Prion Protein Transmembrane Domain Regulate Nascent Chain Orientation J. Biol. Chem., April 13, 2007; 282(15): 11163 - 11171. [Abstract] [Full Text] [PDF]

				P. Clarke and A. C Ghani Projections of the future course of the primary vCJD epidemic in the UK: inclusion of subclinical infection and the possibility of wider genetic susceptibility J R Soc Interface, March 22, 2005; 2(2): 19 - 31. [Abstract] [Full Text] [PDF]

				M. Chadeau-Hyam and A. Alperovitch Risk of variant Creutzfeldt-Jakob disease in France Int. J. Epidemiol., February 1, 2005; 34(1): 46 - 52. [Abstract] [Full Text] [PDF]

				D R Norfolk, A Glaser, and S Kinsey American fresh frozen plasma for neonates and children Arch. Dis. Child., January 1, 2005; 90(1): 89 - 91. [Abstract] [Full Text] [PDF]

				S. M Bird Recipients of blood or blood products "at vCJD risk" BMJ, January 17, 2004; 328(7432): 118 - 119. [Full Text] [PDF]

				G. Davey Smith Uncertainty and significance Int. J. Epidemiol., October 1, 2003; 32(5): 683 - 683. [Full Text] [PDF]

				P. S Brachman Infectious diseases--past, present, and future Int. J. Epidemiol., October 1, 2003; 32(5): 684 - 686. [Full Text] [PDF]

				A. C Ghani Commentary: Predicting the unpredictable: the future incidence of variant Creutzfeldt-Jakob disease Int. J. Epidemiol., October 1, 2003; 32(5): 792 - 793. [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (15) Request Permissions Google Scholar Articles by Cooper, J. Articles by Bird, S. Search for Related Content PubMed PubMed Citation Articles by Cooper, J. Articles by Bird, S. Social Bookmarking          What's this?

Online ISSN 1464-3685 - Print ISSN 0300-5771

Copyright © 2010 International Epidemiological Association

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics